diff --git "a/related_34K/test_related_short_2405.00722v1.json" "b/related_34K/test_related_short_2405.00722v1.json"
new file mode 100644--- /dev/null
+++ "b/related_34K/test_related_short_2405.00722v1.json"
@@ -0,0 +1,1431 @@
+[
+    {
+        "url": "http://arxiv.org/abs/2405.00722v1",
+        "title": "LLMs for Generating and Evaluating Counterfactuals: A Comprehensive Study",
+        "abstract": "As NLP models become more complex, understanding their decisions becomes more\ncrucial. Counterfactuals (CFs), where minimal changes to inputs flip a model's\nprediction, offer a way to explain these models. While Large Language Models\n(LLMs) have shown remarkable performance in NLP tasks, their efficacy in\ngenerating high-quality CFs remains uncertain. This work fills this gap by\ninvestigating how well LLMs generate CFs for two NLU tasks. We conduct a\ncomprehensive comparison of several common LLMs, and evaluate their CFs,\nassessing both intrinsic metrics, and the impact of these CFs on data\naugmentation. Moreover, we analyze differences between human and LLM-generated\nCFs, providing insights for future research directions. Our results show that\nLLMs generate fluent CFs, but struggle to keep the induced changes minimal.\nGenerating CFs for Sentiment Analysis (SA) is less challenging than NLI where\nLLMs show weaknesses in generating CFs that flip the original label. This also\nreflects on the data augmentation performance, where we observe a large gap\nbetween augmenting with human and LLMs CFs. Furthermore, we evaluate LLMs'\nability to assess CFs in a mislabelled data setting, and show that they have a\nstrong bias towards agreeing with the provided labels. GPT4 is more robust\nagainst this bias and its scores correlate well with automatic metrics. Our\nfindings reveal several limitations and point to potential future work\ndirections.",
+        "authors": "Van Bach Nguyen, Paul Youssef, J\u00f6rg Schl\u00f6tterer, Christin Seifert",
+        "published": "2024-04-26",
+        "updated": "2024-04-26",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "label": "Original Paper",
+        "paper_cat": "LLM Fairness",
+        "gt": "Large Language Models. LLMs have demonstrated impressive capabilities across a diverse natural language processing tasks, such as question LLM FL UA RS Avg. SA Human Crowd 3.66 2.95 2.58 3.06 Human Expert 3.54 2.69 2.49 2.91 GPT3.5 3.58 2.91 2.65 3.05 GPT4 3.79 3.15 2.91 3.28 LLAMA2 7B 3.60 2.74 2.63 2.99 LLAMA2 70B 3.70 2.75 2.47 2.97 Mistral 7B 3.85 2.84 2.69 3.13 Mistral 56B 2.58 1.74 1.75 2.02 hypothesis Human Crowd 3.81 3.96 3.81 3.86 GPT3.5 3.19 3.93 3.74 3.62 GPT4 3.96 3.98 3.92 3.95 LLAMA2 7B 3.23 3.74 3.66 3.54 LLAMA2 70B 3.49 3.68 3.56 3.58 Mistral 7B 3.50 3.70 3.65 3.62 Mistral 56B 3.02 3.45 3.48 3.32 premise Human Crowd 3.58 3.88 3.86 3.77 GPT3.5 2.51 3.82 3.69 3.34 GPT4 3.68 3.83 3.84 3.78 LLAMA2 7B 2.96 3.38 3.67 3.34 LLAMA2 70B 3.35 3.46 3.67 3.49 Mistral 7B 2.97 3.63 3.74 3.45 Mistral 56B 2.37 3.11 3.49 2.99 Table 8: Scores for evaluation with GPT4. FL refers to flipping label score, UA to unncessary alteration, RS is the realisticness score, and Avg. is the average of the three scores. Best score for each task is in bold. Second best score is underlined. answering, wherein the model needs to retrieve relevant information from its training data and generate a concise response, or text summarization, which distills lengthy texts into concise summaries while retaining crucial information (Maynez et al., 2023). However, the task of CFs generation has not been comprehensively evaluated for LLMs. A large number of LLMs exist, exhibiting variations in model size, architecture, training dataset, the incorporation of human feedback loops and accessibility (open-source or proprietary) (Zhao et al., 2023). Consequently, there is a necessity to conduct comparative evaluations across different models on a standardized task. Since the architectures of the LLMs under consideration are predominantly similar, and the training datasets are either known Compared Values SA NLI FL & FR 0.83 0.92 UA & -TS 0.50 0.75 RS & -PPL 0.62 -0.23 Table 9: Spearman correlations between intrinsic metrics and GPT-4 evaluation scores. PPL and TS scores are negated so that higher is better. LLM/Score 1 2 3 4 GPT3.5 0.70 3.85 69.61 25.84 GPT4 55.50 3.19 2.94 38.37 Table 10: Flip label score distributions on the corrupted set of NLI. Distribution is an average of the distributions on the premise and hypothesis sets. public sources or undisclosed, the primary focus of this study is to compare LLMs that are different in model size, the implementation of human feedback, and accessibility. To enhance the performance of LLMs across various tasks, in-context learning (ICL) techniques have been employed to optimize the prompts provided to these models. Numerous prompt engineering approaches during the inference phase have been proposed, either by selecting the demonstration instances, or formatting the prompt in form of instruction or reasoning steps (Dong et al., 2022). In this study, leverage chainof-thought prompting (CoT) (Wei et al., 2022) and selecting closest instance retrieval strategies(Liu et al., 2022) to optimize the generation process. CFs generation methods. There exists several methods for generating CFs, but most of them are desigend for a specific LLM. The CFs generated by MICE (Ross et al., 2021) are intended for debugging models, and not for data augmentation. Polyjuice (Wu et al., 2021) requires specifying the type of edits that should be conducted, and the resulting CFs should be manually labeled. (Robeer et al., 2021). DISCO (Chen et al., 2023) uses GPT3\u2019s fill-in-the-blanks mode, which is unavailable in most open source LLMs and would require adapting them. CREST (Treviso et al., 2023) depends on a rationalizer module and the editor module is a masked LM that needs to be further trained. Instead, we decided to prompt LLMs to generate CFs by providing instructions and an example. We provide more details in Section 3.2. LLMs for CFs generation (Li et al., 2024) investigated the strengths and weaknesses of LLMs as CFs generators. Additionally, they disclosed the factors that impact LLMs during CFs generation, including both intrinsic properties of LLMs and prompt design considerations. However, this study lacks intrinsic evaluation of CFs and omits comparison with human-generated CFs. Sachdeva et al. (2024) leverage LLMs to generate CFs for extractive question answering, showing that data augmentation with CFs improve OOD performance, and that this improvement correlates with the diversity of the generated CFs. Prior work by Bhattacharjee et al. (2024) investigated the capability of GPT models in generating CFs for explanations by optimizing their prompts. However, their analysis was limited to the GPT family and did not consider downstream tasks or comparison with humangenerated CFs. In this work, we conduct a more comprehensive evaluation of LLMs on multiple aspects, including intrinsic metrics of CFs explanation quality and performance on downstream tasks. Furthermore, we compare the LLM-generated CFs against those produced by humans, and propose a novel approach to evaluate CFs using LLMs.",
+        "pre_questions": [],
+        "main_content": "Introduction The growing popularity of artificial intelligence (AI) and increasingly complex \u201cblack-box\u201d models have triggered a critical need for interpretability. As Miller (2019) highlights, explanations often seek to understand why an event P occurred instead of an alternative Q. Ideally, explanations \u2217 Equal contribution Positive: If you haven't seen this, it's incredible. It is pure gold. I saw this about 17 years ago, and I'm still hype about it. Positive:\u00a0If you haven't seen this, it's amazing. It is incredible. I saw this about 17 years ago, and I'm still amazed from it. Positive:\u00a0If you haven\u2019t seen this, it\u2019s terrible. It is pure beauty. I saw this about 17 years ago, and I\u2019m still impressed from it. Positive:\u00a0If you haven't seen this, it's amazing. It is a hidden gem. I saw this about 17 years ago, and I'm still enlightened from it. Negative: If you haven't seen this, it's terrible. It is pure trash. I saw this about 17 years ago, and I'm still screwed up from it. Original 3.3 3.0 3.4 3.5 score Figure 1: Counterfactual for Sentiment Analysis from several LLMs with their evaluation scores from GPT4. should demonstrate how minimal changes to an instance could have led to different outcomes. In the context of textual data, this translates to introducing minimal modifications to the text through word additions, replacements, or deletions, to flip the label assigned by a given classifier. Counterfactual generation in NLP aims to foster an understanding of models, thereby facilitating their improvement (Kaushik et al., 2020), debugging (Ross et al., 2021), or rectification (Balashankar et al., 2023). In the field of NLP, LLMs have consistently demonstrated remarkable performance across diverse tasks. However, despite significant advancements in counterfactual generation methods, the efficacy of LLMs in producing high-quality CFs remains an open question. Our study bridges this gap by rigorously assessing the inherent capability of LLMs to generate CFs and identifying the most effective ones. We conduct a comprehensive comparison of several common LLMs, spanning different sizes and accessibility levels, evaluating their performance specifically on the counterfactual generation task. Our assessment encompasses standard metrics for counterfactual quality, as well as an in-depth evaluation of language fluency tailored to the context of counterfactual generation. Furthermore, we extend our analysis to data augmentation. We consider generating CFs for two arXiv:2405.00722v1  [cs.CL]  26 Apr 2024 NLU tasks in this study: Sentiment Analysis (SA) and Natural Language Inference (NLI). Our analysis demonstrates that LLMs are able to generate fluent text. However, they have difficulties in inducing minimal changes. Generating CFs for SA is less challenging than NLI, where LLMs exhibit weaknesses in generating CFs that flip the labels. For data augmentation, CFs from LLMs can be an alternative to human CFs, as they are able to achieve similar performance, while on NLI further improvements are needed. Furthermore, we show a positive correlation between keeping minimal changes and data augmentation performance. This suggests a new direction to generate improved data for augmentation, potentially leading to more efficient augmentation approaches. We further assess the ability of LLMs to act as evaluators of CFs. By conducting controlled experiments, we show that LLMs have a strong bias to agree with the given labels, even if these are incorrect. Additionally, we show the alignment between GPT4-based evaluation and intrinsic metrics for CFs, indicating that GPT-4 is a reliable evaluator for counterfactual generation. These findings suggest that GPT-4 is a suitable choice of LLM to use when evaluating the quality of generated CFs. A sample of CFs from different LLMs with the corresponding scores is shown in Figure 1. Finally, to facilitate further research, we contribute a new dataset of CFs generated by various LLMs. 2 Evaluation Methodology We conduct a multi-faceted evaluation, considering several use cases where CFs could be beneficial. 2.1 Intrinsic Evaluation Given a fixed classifier f and a dataset with N samples (x1, x2, . . . , xN), x = (z1, z2, . . . , zn) represents a sequence of n tokens with a ground truth label y. A valid counterfactual x\u2032 should: (1) achieve the desired target label y\u2032 with (2) minimal changes, and (3) align with likely feature distributions (Molnar, 2022). Therefore, in this evaluation, we consider the intrinsic properties of Flip Rate, Textual Similarity, and Perplexity that correspond to each criterion, respectively: Flip Rate (FR): measures how effectively a method can change labels of instances with respect to a pretrained classifier. FR is defined as the percentage of generated instances where the labels are flipped over the total number of instances N (Bhattacharjee et al., 2024): FR = 1 N N X i=1 1[f(xi) = y\u2032] Textual Similarity (TS): quantifies the closeness between an original instance and the counterfactual. Lower distances indicate greater similarity. We use the Levenshtein distance for d to quantify the tokendistance between the original instance x and the counterfactual x\u2032. This choice is motivated by the Levenshtein distance\u2019s ability to capture all type of edits (insertions, deletions, or substitutions) and also its widespread use in related work (Ross et al., 2021; Treviso et al., 2023): TS = 1 N N X i=1 d(xi, x\u2032 i) |xi| Perplexity (PPL): To ensure that the generated text is plausible, realistic, and follows a natural text distribution, we leverage perplexity from GPT-2 because of its effectiveness in capturing such distributions. (Radford et al., 2019)1 PPL(x) = exp ( \u22121 n n X i=1 log p\u03b8(zi | z<i) ) 2.2 Data Augmentation After detecting failures in task-specific models, CFs can be used to augment the training data, and help close potential flaws in the reasoning of these models (Kaushik et al., 2020). Additionally, data augmentation with CFs increases generalization and OOD performance (Sen et al., 2021; Ding et al., 2024). In this evaluation, we examine how augmenting original training data with human and LLMs-generated CFs reflects on the performance of task-specific models. 2.3 LLMs for CFs Evaluation Evaluation with LLMs has been shown to be a valid alternative to human evaluation on various tasks like open-ended story generation and adversarial attacks (Chiang and Lee, 2023), open-ended questions (Zheng et al., 2023), translation (Kocmi 1While GPT-2 is used for simplicity in this study, any other LLM can be substituted as long as it demonstrates strong text generation capabilities and Federmann, 2023) and natural language generation (Liusie et al., 2024). In this work, we examine how well can LLMs evaluate CFs. Detecting mistakes in CFs with LLMs opens the door for iteratively refining CFs (Madaan et al., 2023). For assessing LLMs in CFs evaluation, we leverage them to evaluate two sets of CFs. An honest set of CFs from humans, and a corrupted set, where we corrupt the ground truth labels. We compare the scores between the two sets and draw conclusions about the realiablity of LLMs for evaluating CFs. 3 Experimental Setup 3.1 Data We compare CFs generated by LLMs against CFs generated by crowd workers (Kaushik et al., 2020) and experts (Gardner et al., 2020) (hereinafter referred to as \u201cHuman Crowd\u201d and \u201cHuman Experts\u201d respectively). Sentiment Analysis (SA). We experiment with the IMDb dataset (Maas et al., 2011). For better comparability, we use the data splits from Kaushik et al. (2020). Natural Language Inference (NLI). We experiment with SNLI (Bowman et al., 2015). Here too, we use the data splits from Kaushik et al. (2020). 3.2 Generating Countefactuals In order to make our study LLMs-focused and computationally feasible, we decided to generate counterfactual in a way that fulfills the following criteria: \u2022 Generated counterfactuals can be used for data augmentation (an evaluation aspect) \u2022 Does not require human intervention (e.g., specifying edits or labeling) \u2022 Does not require additional training in order to make the study computationally feasible \u2022 The resulting counterfactuals should depend only on the evaluated LLM in order to exclude any other confounding factors To create the prompt for the LLMs to generate the counterfactual, we combine two techniques: (1) Selecting the closest factual instance to the current instance (Liu et al., 2022). Since the provided example has a crucial effect on performance (Liu et al., 2022), we select the closest factual/counterfactual pair that has been generated by humans. We use SentenceBERT (Reimers and Gurevych, 2019) to obtain the latent space representation, and then calculate the distance using cosine similarity from that latent space. (2) Chainof-Thought (CoT) prompting (Wei et al., 2022), showing the necessary steps to generate a counterfactual instance based on a factual one, since it has been shown to help LLMs reason better and provide higher-quality answers. An an overview of the process for generating CFs is depicted in Figure 2. Specifically, we use the validation set in each dataset as a reference to select the closest example when generating the counterfactual for the train and test sets. After obtaining the pair of closest instances, we apply CoT prompting by defining three steps to generate the counterfactual: \u2022 Step 1: Identify all of the important words that contribute to flipping the label. \u2022 Step 2: Find replacements for the words identified in Step 1 that lead to the target label. \u2022 Step 3: Replace the words from Step 2 into the original text to obtain the counterfactual instance. This prompt aligns with the principle of other work (Ross et al., 2021; Treviso et al., 2023; Li et al., 2024), which involve identifying significant words that impact the label and altering them to flip the label, thereby generating counterfactual instances. The prompt examples can be found in the Appendix F. If you haven't seen this, it's terrible. It is pure trash. I saw this  about 17 years ago, and I'm still screwed up from it. 1.Identify important words: terrible, pure trash, screwed up 2.Find replacements :  terrible \u2192 amazing pure trash \u2192 hidden gem screwed up \u2192 enlightened  3.Replace the words in the original text If you haven't seen this, it's amazing. It is a hidden gem. I saw  this about 17 years ago, and I'm still enlightened from it. Original  Counterfactual  Instructions on  closest example Figure 2: An overview of CFs generation process. Stepby-step instructions are shown on closest example. 3.3 LLMs We compare open-source LLMs with closed-source LLMs. We choose LLAMA-2 (Touvron et al., 2023) and Mistral (Jiang et al., 2023) as representatives for open-source models, and GPT-3.5 and GPT-42 as a representative for closed-source LLMs. Table 1 summarizes the properties of each LLM. Model Size HF Instruct OS LLAMA2 7B/70B \u2713 \u2713 \u2713 Mistral 7B \u2717 \u2713 \u2713 Mistral 56B \u2717 \u2717 \u2713 GPT3.5 \u2713 \u2713 \u2717 GPT4 \u2713 \u2713 \u2717 Table 1: Characteristics of Large Language Models (LLMs, including Size, Human Feedback (HF), Instruction, and Open-Source (OS). 4 Results and Discussion 4.1 Intrinsic Evaluation We show the results for the intrinsic evaluation in Table 2. For flip rate, we use SOTA BERT-based models from (Morris et al., 2020). The obtained perplexity values reflect the high fluency of LLMs, some of which are even more fluent than humans.3 Distance values show that LLMs do not necessarily adhere to conducting minimal changes. One exception here is GPT3.5, whose average distance values resemble that of humangenerated CFs. The large distance values for LLMsgenerated CFs could be explained by their tendency to overgenerate (Guerreiro et al., 2023). In terms of flip rate, we notice the some LLMsgenerated CFs can have a higher flip rate than human-generated CFs on SA, whereas the opposite can can be observed on NLI. Indeed, NLI CFs could be more difficult to generate than CFs for SA, which explains the gap in flip rate between LLMs and humans on the one hand, and GPT-4 and other LLMs on the other hand (this is especially apparent on the NLI hypothesis test set). This suggests that GPT-4 should be the preferred choice to generate CFs for explaining models behavior. Note that LLAMA2 70B appears to have a higher flip rate on the SA and NLI premise test sets, but it also has a much lower success rate for generating CFs (cf. Table 11 in Appendix A). We examine the LLMs-generated CFs in more detail in Section 4.4. We also notice that Mistral 56B, which has not be trained with instruction-tuning or RLHF, produces low quality CFs in terms of the metrics con2We use API from https://openai.com/ 3Note that the shown perplexity values are based on GPT-2 LLM PPL \u2193 TS\u2193 FR\u2191 SA Expert Humans 51.07 0.16 81.15 Crowd Humans 48.03 0.14 85.66 GPT3.5 49.53 0.16 79.51 GPT4 49.05 0.29 94.03 LLAMA2 7B 46.99 0.64 78.26 LLAMA2 70B 60.46 0.82 96.31 Mistral 7B 48.55 1.06 95.13 Mistral 56B 64.05 1.50 58.55 NLI premise Crowd Humans 74.89 0.17 59.13 GPT3.5 71.62 0.15 35.50 GPT4 73.39 0.28 57.12 LLAMA2 7B 70.34 0.36 41.02 LLAMA2 70B 70.74 0.33 58.20 Mistral 7B 78.34 0.36 37.71 Mistral 56B 75.57 0.34 24.32 NLI hypothesis Crowd Humans 65.67 0.19 79.75 GPT3.5 51.30 0.19 41.50 GPT4 58.35 0.21 65.88 LLAMA2 7B 59.60 0.28 38.64 LLAMA2 70B 47.69 0.32 48.85 Mistral 7B 39.06 0.46 44.11 Mistral 56B 62.77 0.33 37.79 Table 2: Metrics for intrinsic evaluation. PPL is perpelexity using GPT-2. TS is Levenshtein distance. FR is flip rate with respect to a SOTA classifier. sidered here. This highlights the importance of instruction-tuning and RLHF in generating highquality texts. This part of the evaluation shows us that LLMs are able to generate fluent CFs, but struggle to induce minimal changes. It also demonstrates that it is challenging to generate NLI CFs that flip the label, whereas generating SA CFs is less difficult. 4.2 Data Augmentation We train on both original training data and CFs from different LLMs to see if augmenting the training data leads to an improved performance. For comparison, we conduct data augmentation with human CFs as well. The results for SA and NLI are shown in Table 3 and 4 respectively. SA. On the crowd CFs and expert CFs test sets for SA including the LLMs-generated CFs lead to improved performance. Perhaps surprisingly, GPT3.5 and LLAMA2 7B provide the most useful CFs for data augmentation. However, augmenting with human CFs works the best. On the original test set, augmenting with CFs does not improve performance. This shows that the gains in performance from data augmentation are visible only if the test set contains challenging examples. Test Data crowd CFs expert CFs original Original only 92.21 86.27 90.16 Human Crowd 95.90 92.62 89.96 GPT3.5 95.08 91.19 89.96 GPT4 93.65 88.73 90.37 LLAMA2 70B 94.26 90.78 89.55 LLAMA2 7B 95.49 91.19 89.75 Mistral 7B 92.42 88.11 88.93 Mistral 56B 93.85 86.89 84.02 Table 3: Data augmentation results for SA. Classification model is trained on original and LLMs or humangenerated CFs. Evaluation is done on 3 test sets: crowd CFs, expert CFs, original with accuracy as a metric. NLI. On the crowd premise test set of NLI, which consists of CFs that were generated by changing the premise only, we notice that most of the LLMsgenerated CFs help improve the model\u2019s performance by a good margin (> 13 pp). The gap to augmenting with human CFs, however, is still large (\u223c9 pp). On the crowd hyothesis test set, most LLMs lead to a lower performance with two exceptions (GPT4 and LLAMA2 70B), where the performance is on par with no data augmentation. However, here too, there is a large gap to human CFs (\u223c13 pp). On the original test set, augmenting with LLMs-generated CFs hurts performance, while augmenting with human-generated CFs bring good improvements (5 pp). This shows how highquality CFs improve the model\u2019s capabilities, and points to a problem with the LLMs-generated CFs for NLI. Connection with intrinsic metrics. We examine the relation between data augmentation performance on the one hand and perplexity and Levenshtein distance on the other hand. The correlation Test Data crowd Premise crowd Hypothesis original Original only 39.88 58.50 73.50 Human Crowd 65.38 71.25 78.50 GPT3.5 55.62 51.88 55.00 GPT4 55.88 58.63 70.75 LLAMA2 7B 56.25 51.12 57.50 LLAMA2 70B 53.50 58.13 65.75 Mistral 7B 52.88 48.38 57.75 Mistral 56B 43.75 39.75 45.75 Table 4: Data augmentation results for SA. Classification model is trained on original and LLMs or humangenerated CFs with Accuracy metric. values in Table 5 suggest that CFs with lower distance (to the factual instances) bring more improvements. Indeed, classifiers could be insensitive to small changes (Glockner et al., 2018), and having such examples in the training can make classifiers more robust. Having more fluent CFs correlates well with higher performance, but only on SA. SA classifiers could be benefiting from fluent inputs, which are more probable and appear in more contexts. Conversely, the inputs in NLI are short image captions that are not necessarily fluent or grammatical texts. Compared values SA NLI Accuracy & \u2212PPL 0.58 -0.05 Accuracy & \u2212TS 0.86 0.51 Table 5: Spearman correlations between intrinsic metrics and data augmentation performance. In summary, only few LLMs produce CFs that come close to human CFs in terms of performance improvements on SA. On NLI, the results are less positive: LLM CFs bring no improvements on two out of three test sets, and the gap to human CFs is still large. CFs with less changes to the factual instances are more beneficial for data augmentation. 4.3 LLMs for CFs Evaluation We examine how reliable are LLMs for CFs evaluation by asking them to evaluate two sets of human CFs: an honest set and a corrupted set. The \u201chonest set\u201d refers to a collection of human CFs, for which the ground truth labels are provided, whereas the \u201ccorrupted set\u201d consists of instances, for which wrong labels differing from the gold labels are provided. In the context of NLI, the third label, distinct from both the target and factual labels, is selected for inclusion in the corrupted set. For SA, the reverse label is chosen while the factual label remains undisclosed. Initially, we prompt GPT3.5 and GPT4 to assess whether the provided CFs accurately represent the target labels by assigning a score from 1 to 4 (cf. Appendix F). Here, a score of 1 or 2 indicates disagreement (complete or partial) with the target label, while a score of 3 or 4 indicates agreement (partial or complete) with the target label. Ideally, the evaluation LLMs should give high scores to the honest set, and low scores to the corrupted set. We show the distributions for disagreements and agreements in Table 6. On SA, both LLMs perform well, but GPT4 exhibits higher sensitivity to the corrupted examples. On NLI, we notice that GPT3.5 gives high flip label scores to humans CFs with both correct and incorrect labels. GPT4 performs much better, but still exhibits high tendency to agree with wrong labels (\u223c40%). The results can be explained by the tendency of LLMs to agree with the provided answers, especially on reasoning tasks (Zheng et al., 2023). To verify this, we prompt both LLMs to classify the same set of NLI CFs by choosing one the three labels (entailment, neutral, contradiction) using a similar prompt. The classification results in Table 7 show an improved performance compared to asking the same LLMs if they agree with incorrect labels (cf. Table 6). We also compare the flip label score distributions of GPT3.5 and GPT4 on the corrupted set in Table 10, and observe that even though GPT3.5 gives high scores to corrupted inputs it is less certain (most frequent score is 3), whereas GPT4 tends to be more certain and assigns mostly 1 or 4 (> 93%). Evaluation with GPT4. We conduct a widescale CFs evaluation with GPT4. Besides verifying the target label FL, we also ask GPT4 to judge if there are any unnecessary alterations UA, and if the CF is realistic RS. For these aspects, we use a scoring scheme ranging from 1 to 4, where higher scores indicate better performance. The results for GPT-4 evaluation can be found in Table 8. The evaluation scores from GPT4 show that GPT4 CFs are the best overall. On SA, Mistral 7b comes in second place after GPT4, whereas on NLI human CFs take the second position. GPT4 LLM/Set Task 1/2 3/4 Avg. GPT3.5 Honest SA 3.61 96.39 3.43 Corrupted SA 77.42 22.58 1.61 Honest premise 0.63 99.37 3.57 Corrupted premise 5.56 94.44 3.13 Honest hypothesis 1.38 98.62 3.56 Corrupted hypothesis 3.53 96.47 3.28 GPT4 Honest SA 7.53 92.47 3.66 Corrupted SA 98.93 1.08 1.04 Honest premise 12.31 87.69 3.58 Corrupted premise 59.51 40.50 2.19 Honest hypothesis 4.50 95.50 3.81 Corrupted hypothesis 57.87 42.12 2.29 Table 6: Flip label scores distribution for GPT3.5 and GPT4 on honest and corrupted sets. might have a bias to prefer its own generations (Panickssery et al., 2024). The scores for SA are lower, which might be explained by the fact that SA is an easier task than NLI, and therefore evaluating SA CFs is also easier for GPT4, and potentially more accurate. To further verify the evaluation scores from GPT4, we calculate the correlations between GPT4 scores and the scores from the intrinsic evaluation. The correlations shown in Table 9 reflect strong relation on SA with respect to all three aspects (flipping label, minimal changes, realisticness). On NLI, we notice the same pattern except for realisticness/perplexity, where we notice a weak negative relation. This might be due to the nature of NLI instances that are short image captions, and are not necessarily grammatical. LLMs show a high tendency to agree with the provided labels even if these are incorrect, especially on tasks that require reasoning such as NLI. GPT4 evaluation scores correlate with automatic metrics for flipping label, distance and fluency. 4.4 Qualitative Analysis 4.4.1 CFs for NLI We look into a selected set of examples based on the evaluation from GPT4. For each LLM, we pick 2 NLI examples with the highest/lowest scores. We end up with 28 examples. We identify three Set LLM Part Acc. Honest GPT3.5 hypothesis 63.08 Honest GPT3.5 premise 54.90 Honest GPT4 hypothesis 75.75 Honest GPT4 premise 59.25 Table 7: Classification performance on human CFs. Note the improved performance compared to asking LLMs if they agree with a given label (cf. Table 6). categories of errors based on this sample : \u2022 Copy-Paste: When asked to generate a CF, and change the label form contradiction to entailment, LLMs will use the unchanged part (premise or hypothesis) as output. This a clever but lazy way to flip the label to entailment, since two identical sentences would naturally have the label entailment. These CFs were given perfect scores by GPT4. Table 12 in the Appendix shows the percentage of copypaste CFs for all LLMs (at most 4.27% for GPT3.5). \u2022 Negation: When asked to to change the label from entailment to contradiction, LLMs would attempt to negate the premise/hypothesis. The negations does not make sense in the observed CFs, and GPT4 rightly assigns them the lowest scores possible. \u2022 Inconsistency: These examples contain contradictory or illogical sentences. Here, GPT4 sometimes incorrectly assigned these CFs high scores. We show some examples alongside the assigned categories in Table 13 in the Appendix. 4.4.2 Evaluation Scores We also look into the evaluation scores from GPT4 on the same set of examples. We show correct and incorrect evaluations in Table 14. GPT4 assigns high scores to contradictory examples, which partially fulfill the target label, and low scores to examples which contain valid minimal changes. GPT4 could be insensitive to such small changes. In this work, we investigated the use of various LLMs for CFs generation. Our results show that LLMs generate fluent CFs, but struggle to keep the induced changes minimal. Generating CFs for SA is less challenging than NLI, where LLMs show weaknesses in generating CFs that change the original label. CFs from LLMs can replace human CFs for the purpose of data augmentation on SA and achieve similar performance, while on NLI further improvements are needed. Furthermore, our results suggest that CFs with minimal changes are essential for data augmentation. We also showed that when asked to asses CFs, LLMs exhibit a strong bias towards agreeing with the provided label even if this label is incorrect. GPT4 appears to be more robust than GPT3.5 against this bias, and its scores correlate well with automatic metrics. However, some failures still exist as shown in our qualitative analysis, indicating limitations of LLMs that warrant further investigation. Future work should focus on leveraging LLMs for higher quality NLI CFs, which correctly changes the label and keep minimal changes, and assessing the evaluation abilities of LLMs in mislabeled data settings. 7 Limitations We used the default parameters for generating counterfactuals. Experimenting with different parameters might have a non-negligble effect on the results. We included various LLMs in our experiments to be inclusive and be able to compare open-source and closed LLMs. However, these LLMs might have been exposed, during their training, to the data we use from (Kaushik et al., 2020). In this regard, the training data of most open-source and all closed-source LLMs remains unknown. In our qualitative analysis (see Section 4.4), we noticed that GPT4 generated a CF that is identical to a human CF from (Kaushik et al., 2020)."
+    },
+    {
+        "url": "http://arxiv.org/abs/2307.07889v3",
+        "title": "LLM Comparative Assessment: Zero-shot NLG Evaluation through Pairwise Comparisons using Large Language Models",
+        "abstract": "Current developments in large language models (LLMs) have enabled impressive\nzero-shot capabilities across various natural language tasks. An interesting\napplication of these systems is in the automated assessment of natural language\ngeneration (NLG), a highly challenging area with great practical benefit. In\nthis paper, we explore two options for exploiting the emergent abilities of\nLLMs for zero-shot NLG assessment: absolute score prediction, and comparative\nassessment which uses relative comparisons between pairs of candidates. Though\ncomparative assessment has not been extensively studied in NLG assessment, we\nnote that humans often find it more intuitive to compare two options rather\nthan scoring each one independently. This work examines comparative assessment\nfrom multiple perspectives: performance compared to absolute grading;\npositional biases in the prompt; and efficient ranking in terms of the number\nof comparisons. We illustrate that LLM comparative assessment is a simple,\ngeneral and effective approach for NLG assessment. For moderate-sized\nopen-source LLMs, such as FlanT5 and Llama2-chat, comparative assessment is\nsuperior to prompt scoring, and in many cases can achieve performance\ncompetitive with state-of-the-art methods. Additionally, we demonstrate that\nLLMs often exhibit strong positional biases when making pairwise comparisons,\nand we propose debiasing methods that can further improve performance.",
+        "authors": "Adian Liusie, Potsawee Manakul, Mark J. F. Gales",
+        "published": "2023-07-15",
+        "updated": "2024-02-06",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2012.13985v2",
+        "title": "Explaining NLP Models via Minimal Contrastive Editing (MiCE)",
+        "abstract": "Humans have been shown to give contrastive explanations, which explain why an\nobserved event happened rather than some other counterfactual event (the\ncontrast case). Despite the influential role that contrastivity plays in how\nhumans explain, this property is largely missing from current methods for\nexplaining NLP models. We present Minimal Contrastive Editing (MiCE), a method\nfor producing contrastive explanations of model predictions in the form of\nedits to inputs that change model outputs to the contrast case. Our experiments\nacross three tasks--binary sentiment classification, topic classification, and\nmultiple-choice question answering--show that MiCE is able to produce edits\nthat are not only contrastive, but also minimal and fluent, consistent with\nhuman contrastive edits. We demonstrate how MiCE edits can be used for two use\ncases in NLP system development--debugging incorrect model outputs and\nuncovering dataset artifacts--and thereby illustrate that producing contrastive\nexplanations is a promising research direction for model interpretability.",
+        "authors": "Alexis Ross, Ana Marasovi\u0107, Matthew E. Peters",
+        "published": "2020-12-27",
+        "updated": "2021-06-23",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2309.13340v2",
+        "title": "Towards LLM-guided Causal Explainability for Black-box Text Classifiers",
+        "abstract": "With the advent of larger and more complex deep learning models, such as in\nNatural Language Processing (NLP), model qualities like explainability and\ninterpretability, albeit highly desirable, are becoming harder challenges to\ntackle and solve. For example, state-of-the-art models in text classification\nare black-box by design. Although standard explanation methods provide some\ndegree of explainability, these are mostly correlation-based methods and do not\nprovide much insight into the model. The alternative of causal explainability\nis more desirable to achieve but extremely challenging in NLP due to a variety\nof reasons. Inspired by recent endeavors to utilize Large Language Models\n(LLMs) as experts, in this work, we aim to leverage the instruction-following\nand textual understanding capabilities of recent state-of-the-art LLMs to\nfacilitate causal explainability via counterfactual explanation generation for\nblack-box text classifiers. To do this, we propose a three-step pipeline via\nwhich, we use an off-the-shelf LLM to: (1) identify the latent or unobserved\nfeatures in the input text, (2) identify the input features associated with the\nlatent features, and finally (3) use the identified input features to generate\na counterfactual explanation. We experiment with our pipeline on multiple NLP\ntext classification datasets, with several recent LLMs, and present interesting\nand promising findings.",
+        "authors": "Amrita Bhattacharjee, Raha Moraffah, Joshua Garland, Huan Liu",
+        "published": "2023-09-23",
+        "updated": "2024-01-29",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.LG"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.17075v1",
+        "title": "CREST: A Joint Framework for Rationalization and Counterfactual Text Generation",
+        "abstract": "Selective rationales and counterfactual examples have emerged as two\neffective, complementary classes of interpretability methods for analyzing and\ntraining NLP models. However, prior work has not explored how these methods can\nbe integrated to combine their complementary advantages. We overcome this\nlimitation by introducing CREST (ContRastive Edits with Sparse\nraTionalization), a joint framework for selective rationalization and\ncounterfactual text generation, and show that this framework leads to\nimprovements in counterfactual quality, model robustness, and interpretability.\nFirst, CREST generates valid counterfactuals that are more natural than those\nproduced by previous methods, and subsequently can be used for data\naugmentation at scale, reducing the need for human-generated examples. Second,\nwe introduce a new loss function that leverages CREST counterfactuals to\nregularize selective rationales and show that this regularization improves both\nmodel robustness and rationale quality, compared to methods that do not\nleverage CREST counterfactuals. Our results demonstrate that CREST successfully\nbridges the gap between selective rationales and counterfactual examples,\naddressing the limitations of existing methods and providing a more\ncomprehensive view of a model's predictions.",
+        "authors": "Marcos Treviso, Alexis Ross, Nuno M. Guerreiro, Andr\u00e9 F. T. Martins",
+        "published": "2023-05-26",
+        "updated": "2023-05-26",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2309.07822v3",
+        "title": "CATfOOD: Counterfactual Augmented Training for Improving Out-of-Domain Performance and Calibration",
+        "abstract": "In recent years, large language models (LLMs) have shown remarkable\ncapabilities at scale, particularly at generating text conditioned on a prompt.\nIn our work, we investigate the use of LLMs to augment training data of small\nlanguage models~(SLMs) with automatically generated counterfactual~(CF)\ninstances -- i.e. minimally altered inputs -- in order to improve\nout-of-domain~(OOD) performance of SLMs in the extractive question\nanswering~(QA) setup. We show that, across various LLM generators, such data\naugmentation consistently enhances OOD performance and improves model\ncalibration for both confidence-based and rationale-augmented calibrator\nmodels. Furthermore, these performance improvements correlate with higher\ndiversity of CF instances in terms of their surface form and semantic content.\nFinally, we show that CF augmented models which are easier to calibrate also\nexhibit much lower entropy when assigning importance, indicating that\nrationale-augmented calibrators prefer concise explanations.",
+        "authors": "Rachneet Sachdeva, Martin Tutek, Iryna Gurevych",
+        "published": "2023-09-14",
+        "updated": "2024-02-13",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2306.16793v1",
+        "title": "Benchmarking Large Language Model Capabilities for Conditional Generation",
+        "abstract": "Pre-trained large language models (PLMs) underlie most new developments in\nnatural language processing. They have shifted the field from\napplication-specific model pipelines to a single model that is adapted to a\nwide range of tasks. Autoregressive PLMs like GPT-3 or PaLM, alongside\ntechniques like few-shot learning, have additionally shifted the output\nmodality to generation instead of classification or regression. Despite their\nubiquitous use, the generation quality of language models is rarely evaluated\nwhen these models are introduced. Additionally, it is unclear how existing\ngeneration tasks--while they can be used to compare systems at a high\nlevel--relate to the real world use cases for which people have been adopting\nthem. In this work, we discuss how to adapt existing application-specific\ngeneration benchmarks to PLMs and provide an in-depth, empirical study of the\nlimitations and capabilities of PLMs in natural language generation tasks along\ndimensions such as scale, architecture, input and output language. Our results\nshow that PLMs differ in their applicability to different data regimes and\ntheir generalization to multiple languages and inform which PLMs to use for a\ngiven generation task setup. We share best practices to be taken into\nconsideration when benchmarking generation capabilities during the development\nof upcoming PLMs.",
+        "authors": "Joshua Maynez, Priyanka Agrawal, Sebastian Gehrmann",
+        "published": "2023-06-29",
+        "updated": "2023-06-29",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2101.00288v2",
+        "title": "Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Models",
+        "abstract": "While counterfactual examples are useful for analysis and training of NLP\nmodels, current generation methods either rely on manual labor to create very\nfew counterfactuals, or only instantiate limited types of perturbations such as\nparaphrases or word substitutions. We present Polyjuice, a general-purpose\ncounterfactual generator that allows for control over perturbation types and\nlocations, trained by finetuning GPT-2 on multiple datasets of paired\nsentences. We show that Polyjuice produces diverse sets of realistic\ncounterfactuals, which in turn are useful in various distinct applications:\nimproving training and evaluation on three different tasks (with around 70%\nless annotation effort than manual generation), augmenting state-of-the-art\nexplanation techniques, and supporting systematic counterfactual error analysis\nby revealing behaviors easily missed by human experts.",
+        "authors": "Tongshuang Wu, Marco Tulio Ribeiro, Jeffrey Heer, Daniel S. Weld",
+        "published": "2021-01-01",
+        "updated": "2021-06-01",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2212.10534v3",
+        "title": "DISCO: Distilling Counterfactuals with Large Language Models",
+        "abstract": "Models trained with counterfactually augmented data learn representations of\nthe causal structure of tasks, enabling robust generalization. However,\nhigh-quality counterfactual data is scarce for most tasks and not easily\ngenerated at scale. When crowdsourced, such data is typically limited in scale\nand diversity; when generated using supervised methods, it is computationally\nexpensive to extend to new counterfactual dimensions. In this work, we\nintroduce DISCO (DIStilled COunterfactual Data), a new method for automatically\ngenerating high quality counterfactual data at scale. DISCO engineers prompts\nto generate phrasal perturbations with a large general language model. Then, a\ntask-specific teacher model filters these generations to distill high-quality\ncounterfactual data. While task-agnostic, we apply our pipeline to the task of\nnatural language inference (NLI) and find that on challenging evaluations such\nas the NLI stress test, comparatively smaller student models trained with DISCO\ngenerated counterfactuals are more robust (6% absolute) and generalize better\nacross distributions (2%) compared to models trained without data augmentation.\nFurthermore, DISCO augmented models are 10% more consistent between\ncounterfactual pairs on three evaluation sets, demonstrating that DISCO\naugmentation enables models to more reliably learn causal representations. Our\nrepository is available at: https://github.com/eric11eca/disco",
+        "authors": "Zeming Chen, Qiyue Gao, Antoine Bosselut, Ashish Sabharwal, Kyle Richardson",
+        "published": "2022-12-20",
+        "updated": "2023-06-05",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "label": "Related Work"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.14607v2",
+        "title": "Confronting LLMs with Traditional ML: Rethinking the Fairness of Large Language Models in Tabular Classifications",
+        "abstract": "Recent literature has suggested the potential of using large language models\n(LLMs) to make classifications for tabular tasks. However, LLMs have been shown\nto exhibit harmful social biases that reflect the stereotypes and inequalities\npresent in society. To this end, as well as the widespread use of tabular data\nin many high-stake applications, it is important to explore the following\nquestions: what sources of information do LLMs draw upon when making\nclassifications for tabular tasks; whether and to what extent are LLM\nclassifications for tabular data influenced by social biases and stereotypes;\nand what are the consequential implications for fairness?\n  Through a series of experiments, we delve into these questions and show that\nLLMs tend to inherit social biases from their training data which significantly\nimpact their fairness in tabular classification tasks. Furthermore, our\ninvestigations show that in the context of bias mitigation, though in-context\nlearning and finetuning have a moderate effect, the fairness metric gap between\ndifferent subgroups is still larger than that in traditional machine learning\nmodels, such as Random Forest and shallow Neural Networks. This observation\nemphasizes that the social biases are inherent within the LLMs themselves and\ninherited from their pretraining corpus, not only from the downstream task\ndatasets. Besides, we demonstrate that label-flipping of in-context examples\ncan significantly reduce biases, further highlighting the presence of inherent\nbias within LLMs.",
+        "authors": "Yanchen Liu, Srishti Gautam, Jiaqi Ma, Himabindu Lakkaraju",
+        "published": "2023-10-23",
+        "updated": "2024-04-02",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.07884v2",
+        "title": "Fair Abstractive Summarization of Diverse Perspectives",
+        "abstract": "People from different social and demographic groups express diverse\nperspectives and conflicting opinions on a broad set of topics such as product\nreviews, healthcare, law, and politics. A fair summary should provide a\ncomprehensive coverage of diverse perspectives without underrepresenting\ncertain groups. However, current work in summarization metrics and Large\nLanguage Models (LLMs) evaluation has not explored fair abstractive\nsummarization. In this paper, we systematically investigate fair abstractive\nsummarization for user-generated data. We first formally define fairness in\nabstractive summarization as not underrepresenting perspectives of any groups\nof people, and we propose four reference-free automatic metrics by measuring\nthe differences between target and source perspectives. We evaluate nine LLMs,\nincluding three GPT models, four LLaMA models, PaLM 2, and Claude, on six\ndatasets collected from social media, online reviews, and recorded transcripts.\nExperiments show that both the model-generated and the human-written reference\nsummaries suffer from low fairness. We conduct a comprehensive analysis of the\ncommon factors influencing fairness and propose three simple but effective\nmethods to alleviate unfair summarization. Our dataset and code are available\nat https://github.com/psunlpgroup/FairSumm.",
+        "authors": "Yusen Zhang, Nan Zhang, Yixin Liu, Alexander Fabbri, Junru Liu, Ryo Kamoi, Xiaoxin Lu, Caiming Xiong, Jieyu Zhao, Dragomir Radev, Kathleen McKeown, Rui Zhang",
+        "published": "2023-11-14",
+        "updated": "2024-03-30",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.02839v1",
+        "title": "An Empirical Study of LLM-as-a-Judge for LLM Evaluation: Fine-tuned Judge Models are Task-specific Classifiers",
+        "abstract": "Recently, there has been a growing trend of utilizing Large Language Model\n(LLM) to evaluate the quality of other LLMs. Many studies have employed\nproprietary close-source models, especially GPT4, as the evaluator.\nAlternatively, other works have fine-tuned judge models based on open-source\nLLMs as the evaluator. In this study, we conduct an empirical study of\ndifferent judge models on their evaluation capability. Our findings indicate\nthat although the fine-tuned judge models achieve high accuracy on in-domain\ntest sets, even surpassing GPT4, they are inherently task-specific classifiers,\nand their generalizability and fairness severely underperform GPT4.",
+        "authors": "Hui Huang, Yingqi Qu, Jing Liu, Muyun Yang, Tiejun Zhao",
+        "published": "2024-03-05",
+        "updated": "2024-03-05",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2307.03838v2",
+        "title": "RADAR: Robust AI-Text Detection via Adversarial Learning",
+        "abstract": "Recent advances in large language models (LLMs) and the intensifying\npopularity of ChatGPT-like applications have blurred the boundary of\nhigh-quality text generation between humans and machines. However, in addition\nto the anticipated revolutionary changes to our technology and society, the\ndifficulty of distinguishing LLM-generated texts (AI-text) from human-generated\ntexts poses new challenges of misuse and fairness, such as fake content\ngeneration, plagiarism, and false accusations of innocent writers. While\nexisting works show that current AI-text detectors are not robust to LLM-based\nparaphrasing, this paper aims to bridge this gap by proposing a new framework\ncalled RADAR, which jointly trains a robust AI-text detector via adversarial\nlearning. RADAR is based on adversarial training of a paraphraser and a\ndetector. The paraphraser's goal is to generate realistic content to evade\nAI-text detection. RADAR uses the feedback from the detector to update the\nparaphraser, and vice versa. Evaluated with 8 different LLMs (Pythia, Dolly\n2.0, Palmyra, Camel, GPT-J, Dolly 1.0, LLaMA, and Vicuna) across 4 datasets,\nexperimental results show that RADAR significantly outperforms existing AI-text\ndetection methods, especially when paraphrasing is in place. We also identify\nthe strong transferability of RADAR from instruction-tuned LLMs to other LLMs,\nand evaluate the improved capability of RADAR via GPT-3.5-Turbo.",
+        "authors": "Xiaomeng Hu, Pin-Yu Chen, Tsung-Yi Ho",
+        "published": "2023-07-07",
+        "updated": "2023-10-24",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2401.01262v2",
+        "title": "Fairness Certification for Natural Language Processing and Large Language Models",
+        "abstract": "Natural Language Processing (NLP) plays an important role in our daily lives,\nparticularly due to the enormous progress of Large Language Models (LLM).\nHowever, NLP has many fairness-critical use cases, e.g., as an expert system in\nrecruitment or as an LLM-based tutor in education. Since NLP is based on human\nlanguage, potentially harmful biases can diffuse into NLP systems and produce\nunfair results, discriminate against minorities or generate legal issues.\nHence, it is important to develop a fairness certification for NLP approaches.\nWe follow a qualitative research approach towards a fairness certification for\nNLP. In particular, we have reviewed a large body of literature on algorithmic\nfairness, and we have conducted semi-structured expert interviews with a wide\nrange of experts from that area. We have systematically devised six fairness\ncriteria for NLP, which can be further refined into 18 sub-categories. Our\ncriteria offer a foundation for operationalizing and testing processes to\ncertify fairness, both from the perspective of the auditor and the audited\norganization.",
+        "authors": "Vincent Freiberger, Erik Buchmann",
+        "published": "2024-01-02",
+        "updated": "2024-01-03",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.CY",
+            "cs.LG",
+            "68T50",
+            "I.2.7"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.18569v1",
+        "title": "Fairness of ChatGPT",
+        "abstract": "Understanding and addressing unfairness in LLMs are crucial for responsible\nAI deployment. However, there is a limited availability of quantitative\nanalyses and in-depth studies regarding fairness evaluations in LLMs,\nespecially when applying LLMs to high-stakes fields. This work aims to fill\nthis gap by providing a systematic evaluation of the effectiveness and fairness\nof LLMs using ChatGPT as a study case. We focus on assessing ChatGPT's\nperformance in high-takes fields including education, criminology, finance and\nhealthcare. To make thorough evaluation, we consider both group fairness and\nindividual fairness and we also observe the disparities in ChatGPT's outputs\nunder a set of biased or unbiased prompts. This work contributes to a deeper\nunderstanding of LLMs' fairness performance, facilitates bias mitigation and\nfosters the development of responsible artificial intelligence systems.",
+        "authors": "Yunqi Li, Yongfeng Zhang",
+        "published": "2023-05-22",
+        "updated": "2023-05-22",
+        "primary_cat": "cs.LG",
+        "cats": [
+            "cs.LG",
+            "cs.AI",
+            "cs.CL",
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.18333v3",
+        "title": "She had Cobalt Blue Eyes: Prompt Testing to Create Aligned and Sustainable Language Models",
+        "abstract": "As the use of large language models (LLMs) increases within society, as does\nthe risk of their misuse. Appropriate safeguards must be in place to ensure LLM\noutputs uphold the ethical standards of society, highlighting the positive role\nthat artificial intelligence technologies can have. Recent events indicate\nethical concerns around conventionally trained LLMs, leading to overall unsafe\nuser experiences. This motivates our research question: how do we ensure LLM\nalignment? In this work, we introduce a test suite of unique prompts to foster\nthe development of aligned LLMs that are fair, safe, and robust. We show that\nprompting LLMs at every step of the development pipeline, including data\ncuration, pre-training, and fine-tuning, will result in an overall more\nresponsible model. Our test suite evaluates outputs from four state-of-the-art\nlanguage models: GPT-3.5, GPT-4, OPT, and LLaMA-2. The assessment presented in\nthis paper highlights a gap between societal alignment and the capabilities of\ncurrent LLMs. Additionally, implementing a test suite such as ours lowers the\nenvironmental overhead of making models safe and fair.",
+        "authors": "Veronica Chatrath, Oluwanifemi Bamgbose, Shaina Raza",
+        "published": "2023-10-20",
+        "updated": "2023-12-15",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.18502v1",
+        "title": "Few-Shot Fairness: Unveiling LLM's Potential for Fairness-Aware Classification",
+        "abstract": "Employing Large Language Models (LLM) in various downstream applications such\nas classification is crucial, especially for smaller companies lacking the\nexpertise and resources required for fine-tuning a model. Fairness in LLMs\nhelps ensure inclusivity, equal representation based on factors such as race,\ngender and promotes responsible AI deployment. As the use of LLMs has become\nincreasingly prevalent, it is essential to assess whether LLMs can generate\nfair outcomes when subjected to considerations of fairness. In this study, we\nintroduce a framework outlining fairness regulations aligned with various\nfairness definitions, with each definition being modulated by varying degrees\nof abstraction. We explore the configuration for in-context learning and the\nprocedure for selecting in-context demonstrations using RAG, while\nincorporating fairness rules into the process. Experiments conducted with\ndifferent LLMs indicate that GPT-4 delivers superior results in terms of both\naccuracy and fairness compared to other models. This work is one of the early\nattempts to achieve fairness in prediction tasks by utilizing LLMs through\nin-context learning.",
+        "authors": "Garima Chhikara, Anurag Sharma, Kripabandhu Ghosh, Abhijnan Chakraborty",
+        "published": "2024-02-28",
+        "updated": "2024-02-28",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2309.08836v2",
+        "title": "Bias and Fairness in Chatbots: An Overview",
+        "abstract": "Chatbots have been studied for more than half a century. With the rapid\ndevelopment of natural language processing (NLP) technologies in recent years,\nchatbots using large language models (LLMs) have received much attention\nnowadays. Compared with traditional ones, modern chatbots are more powerful and\nhave been used in real-world applications. There are however, bias and fairness\nconcerns in modern chatbot design. Due to the huge amounts of training data,\nextremely large model sizes, and lack of interpretability, bias mitigation and\nfairness preservation of modern chatbots are challenging. Thus, a comprehensive\noverview on bias and fairness in chatbot systems is given in this paper. The\nhistory of chatbots and their categories are first reviewed. Then, bias sources\nand potential harms in applications are analyzed. Considerations in designing\nfair and unbiased chatbot systems are examined. Finally, future research\ndirections are discussed.",
+        "authors": "Jintang Xue, Yun-Cheng Wang, Chengwei Wei, Xiaofeng Liu, Jonghye Woo, C. -C. Jay Kuo",
+        "published": "2023-09-16",
+        "updated": "2023-12-10",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.14208v2",
+        "title": "Content Conditional Debiasing for Fair Text Embedding",
+        "abstract": "Mitigating biases in machine learning models has gained increasing attention\nin Natural Language Processing (NLP). Yet, only a few studies focus on fair\ntext embeddings, which are crucial yet challenging for real-world applications.\nIn this paper, we propose a novel method for learning fair text embeddings. We\nachieve fairness while maintaining utility trade-off by ensuring conditional\nindependence between sensitive attributes and text embeddings conditioned on\nthe content. Specifically, we enforce that embeddings of texts with different\nsensitive attributes but identical content maintain the same distance toward\nthe embedding of their corresponding neutral text. Furthermore, we address the\nissue of lacking proper training data by using Large Language Models (LLMs) to\naugment texts into different sensitive groups. Our extensive evaluations\ndemonstrate that our approach effectively improves fairness while preserving\nthe utility of embeddings, representing a pioneering effort in achieving\nconditional independence for fair text embeddings.",
+        "authors": "Wenlong Deng, Blair Chen, Xiaoxiao Li, Christos Thrampoulidis",
+        "published": "2024-02-22",
+        "updated": "2024-02-23",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.CY",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.08780v1",
+        "title": "\"Im not Racist but...\": Discovering Bias in the Internal Knowledge of Large Language Models",
+        "abstract": "Large language models (LLMs) have garnered significant attention for their\nremarkable performance in a continuously expanding set of natural language\nprocessing tasks. However, these models have been shown to harbor inherent\nsocietal biases, or stereotypes, which can adversely affect their performance\nin their many downstream applications. In this paper, we introduce a novel,\npurely prompt-based approach to uncover hidden stereotypes within any arbitrary\nLLM. Our approach dynamically generates a knowledge representation of internal\nstereotypes, enabling the identification of biases encoded within the LLM's\ninternal knowledge. By illuminating the biases present in LLMs and offering a\nsystematic methodology for their analysis, our work contributes to advancing\ntransparency and promoting fairness in natural language processing systems.",
+        "authors": "Abel Salinas, Louis Penafiel, Robert McCormack, Fred Morstatter",
+        "published": "2023-10-13",
+        "updated": "2023-10-13",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2308.05374v2",
+        "title": "Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment",
+        "abstract": "Ensuring alignment, which refers to making models behave in accordance with\nhuman intentions [1,2], has become a critical task before deploying large\nlanguage models (LLMs) in real-world applications. For instance, OpenAI devoted\nsix months to iteratively aligning GPT-4 before its release [3]. However, a\nmajor challenge faced by practitioners is the lack of clear guidance on\nevaluating whether LLM outputs align with social norms, values, and\nregulations. This obstacle hinders systematic iteration and deployment of LLMs.\nTo address this issue, this paper presents a comprehensive survey of key\ndimensions that are crucial to consider when assessing LLM trustworthiness. The\nsurvey covers seven major categories of LLM trustworthiness: reliability,\nsafety, fairness, resistance to misuse, explainability and reasoning, adherence\nto social norms, and robustness. Each major category is further divided into\nseveral sub-categories, resulting in a total of 29 sub-categories.\nAdditionally, a subset of 8 sub-categories is selected for further\ninvestigation, where corresponding measurement studies are designed and\nconducted on several widely-used LLMs. The measurement results indicate that,\nin general, more aligned models tend to perform better in terms of overall\ntrustworthiness. However, the effectiveness of alignment varies across the\ndifferent trustworthiness categories considered. This highlights the importance\nof conducting more fine-grained analyses, testing, and making continuous\nimprovements on LLM alignment. By shedding light on these key dimensions of LLM\ntrustworthiness, this paper aims to provide valuable insights and guidance to\npractitioners in the field. Understanding and addressing these concerns will be\ncrucial in achieving reliable and ethically sound deployment of LLMs in various\napplications.",
+        "authors": "Yang Liu, Yuanshun Yao, Jean-Francois Ton, Xiaoying Zhang, Ruocheng Guo, Hao Cheng, Yegor Klochkov, Muhammad Faaiz Taufiq, Hang Li",
+        "published": "2023-08-10",
+        "updated": "2024-03-21",
+        "primary_cat": "cs.AI",
+        "cats": [
+            "cs.AI",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.09219v5",
+        "title": "\"Kelly is a Warm Person, Joseph is a Role Model\": Gender Biases in LLM-Generated Reference Letters",
+        "abstract": "Large Language Models (LLMs) have recently emerged as an effective tool to\nassist individuals in writing various types of content, including professional\ndocuments such as recommendation letters. Though bringing convenience, this\napplication also introduces unprecedented fairness concerns. Model-generated\nreference letters might be directly used by users in professional scenarios. If\nunderlying biases exist in these model-constructed letters, using them without\nscrutinization could lead to direct societal harms, such as sabotaging\napplication success rates for female applicants. In light of this pressing\nissue, it is imminent and necessary to comprehensively study fairness issues\nand associated harms in this real-world use case. In this paper, we critically\nexamine gender biases in LLM-generated reference letters. Drawing inspiration\nfrom social science findings, we design evaluation methods to manifest biases\nthrough 2 dimensions: (1) biases in language style and (2) biases in lexical\ncontent. We further investigate the extent of bias propagation by analyzing the\nhallucination bias of models, a term that we define to be bias exacerbation in\nmodel-hallucinated contents. Through benchmarking evaluation on 2 popular LLMs-\nChatGPT and Alpaca, we reveal significant gender biases in LLM-generated\nrecommendation letters. Our findings not only warn against using LLMs for this\napplication without scrutinization, but also illuminate the importance of\nthoroughly studying hidden biases and harms in LLM-generated professional\ndocuments.",
+        "authors": "Yixin Wan, George Pu, Jiao Sun, Aparna Garimella, Kai-Wei Chang, Nanyun Peng",
+        "published": "2023-10-13",
+        "updated": "2023-12-01",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.05668v1",
+        "title": "CFaiRLLM: Consumer Fairness Evaluation in Large-Language Model Recommender System",
+        "abstract": "In the evolving landscape of recommender systems, the integration of Large\nLanguage Models (LLMs) such as ChatGPT marks a new era, introducing the concept\nof Recommendation via LLM (RecLLM). While these advancements promise\nunprecedented personalization and efficiency, they also bring to the fore\ncritical concerns regarding fairness, particularly in how recommendations might\ninadvertently perpetuate or amplify biases associated with sensitive user\nattributes. In order to address these concerns, our study introduces a\ncomprehensive evaluation framework, CFaiRLLM, aimed at evaluating (and thereby\nmitigating) biases on the consumer side within RecLLMs.\n  Our research methodically assesses the fairness of RecLLMs by examining how\nrecommendations might vary with the inclusion of sensitive attributes such as\ngender, age, and their intersections, through both similarity alignment and\ntrue preference alignment. By analyzing recommendations generated under\ndifferent conditions-including the use of sensitive attributes in user\nprompts-our framework identifies potential biases in the recommendations\nprovided. A key part of our study involves exploring how different detailed\nstrategies for constructing user profiles (random, top-rated, recent) impact\nthe alignment between recommendations made without consideration of sensitive\nattributes and those that are sensitive-attribute-aware, highlighting the bias\nmechanisms within RecLLMs.\n  The findings in our study highlight notable disparities in the fairness of\nrecommendations, particularly when sensitive attributes are integrated into the\nrecommendation process, either individually or in combination. The analysis\ndemonstrates that the choice of user profile sampling strategy plays a\nsignificant role in affecting fairness outcomes, highlighting the complexity of\nachieving fair recommendations in the era of LLMs.",
+        "authors": "Yashar Deldjoo, Tommaso di Noia",
+        "published": "2024-03-08",
+        "updated": "2024-03-08",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.18276v1",
+        "title": "Bias Neutralization Framework: Measuring Fairness in Large Language Models with Bias Intelligence Quotient (BiQ)",
+        "abstract": "The burgeoning influence of Large Language Models (LLMs) in shaping public\ndiscourse and decision-making underscores the imperative to address inherent\nbiases within these AI systems. In the wake of AI's expansive integration\nacross sectors, addressing racial bias in LLMs has never been more critical.\nThis paper introduces a novel framework called Comprehensive Bias\nNeutralization Framework (CBNF) which embodies an innovative approach to\nquantifying and mitigating biases within LLMs. Our framework combines the Large\nLanguage Model Bias Index (LLMBI) [Oketunji, A., Anas, M., Saina, D., (2023)]\nand Bias removaL with No Demographics (BLIND) [Orgad, H., Belinkov, Y. (2023)]\nmethodologies to create a new metric called Bias Intelligence Quotient\n(BiQ)which detects, measures, and mitigates racial bias in LLMs without\nreliance on demographic annotations.\n  By introducing a new metric called BiQ that enhances LLMBI with additional\nfairness metrics, CBNF offers a multi-dimensional metric for bias assessment,\nunderscoring the necessity of a nuanced approach to fairness in AI [Mehrabi et\nal., 2021]. This paper presents a detailed analysis of Latimer AI (a language\nmodel incrementally trained on black history and culture) in comparison to\nChatGPT 3.5, illustrating Latimer AI's efficacy in detecting racial, cultural,\nand gender biases through targeted training and refined bias mitigation\nstrategies [Latimer & Bender, 2023].",
+        "authors": "Malur Narayan, John Pasmore, Elton Sampaio, Vijay Raghavan, Gabriella Waters",
+        "published": "2024-04-28",
+        "updated": "2024-04-28",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "D.1; I.2"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2401.15585v1",
+        "title": "Evaluating Gender Bias in Large Language Models via Chain-of-Thought Prompting",
+        "abstract": "There exist both scalable tasks, like reading comprehension and\nfact-checking, where model performance improves with model size, and unscalable\ntasks, like arithmetic reasoning and symbolic reasoning, where model\nperformance does not necessarily improve with model size. Large language models\n(LLMs) equipped with Chain-of-Thought (CoT) prompting are able to make accurate\nincremental predictions even on unscalable tasks. Unfortunately, despite their\nexceptional reasoning abilities, LLMs tend to internalize and reproduce\ndiscriminatory societal biases. Whether CoT can provide discriminatory or\negalitarian rationalizations for the implicit information in unscalable tasks\nremains an open question.\n  In this study, we examine the impact of LLMs' step-by-step predictions on\ngender bias in unscalable tasks. For this purpose, we construct a benchmark for\nan unscalable task where the LLM is given a list of words comprising feminine,\nmasculine, and gendered occupational words, and is required to count the number\nof feminine and masculine words. In our CoT prompts, we require the LLM to\nexplicitly indicate whether each word in the word list is a feminine or\nmasculine before making the final predictions. With counting and handling the\nmeaning of words, this benchmark has characteristics of both arithmetic\nreasoning and symbolic reasoning. Experimental results in English show that\nwithout step-by-step prediction, most LLMs make socially biased predictions,\ndespite the task being as simple as counting words. Interestingly, CoT\nprompting reduces this unconscious social bias in LLMs and encourages fair\npredictions.",
+        "authors": "Masahiro Kaneko, Danushka Bollegala, Naoaki Okazaki, Timothy Baldwin",
+        "published": "2024-01-28",
+        "updated": "2024-01-28",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.00811v1",
+        "title": "Cognitive Bias in High-Stakes Decision-Making with LLMs",
+        "abstract": "Large language models (LLMs) offer significant potential as tools to support\nan expanding range of decision-making tasks. However, given their training on\nhuman (created) data, LLMs can inherit both societal biases against protected\ngroups, as well as be subject to cognitive bias. Such human-like bias can\nimpede fair and explainable decisions made with LLM assistance. Our work\nintroduces BiasBuster, a framework designed to uncover, evaluate, and mitigate\ncognitive bias in LLMs, particularly in high-stakes decision-making tasks.\nInspired by prior research in psychology and cognitive sciences, we develop a\ndataset containing 16,800 prompts to evaluate different cognitive biases (e.g.,\nprompt-induced, sequential, inherent). We test various bias mitigation\nstrategies, amidst proposing a novel method using LLMs to debias their own\nprompts. Our analysis provides a comprehensive picture on the presence and\neffects of cognitive bias across different commercial and open-source models.\nWe demonstrate that our self-help debiasing effectively mitigate cognitive bias\nwithout having to manually craft examples for each bias type.",
+        "authors": "Jessica Echterhoff, Yao Liu, Abeer Alessa, Julian McAuley, Zexue He",
+        "published": "2024-02-25",
+        "updated": "2024-02-25",
+        "primary_cat": "cs.AI",
+        "cats": [
+            "cs.AI",
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.09447v2",
+        "title": "How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilities",
+        "abstract": "The rapid progress in open-source Large Language Models (LLMs) is\nsignificantly driving AI development forward. However, there is still a limited\nunderstanding of their trustworthiness. Deploying these models at scale without\nsufficient trustworthiness can pose significant risks, highlighting the need to\nuncover these issues promptly. In this work, we conduct an adversarial\nassessment of open-source LLMs on trustworthiness, scrutinizing them across\neight different aspects including toxicity, stereotypes, ethics, hallucination,\nfairness, sycophancy, privacy, and robustness against adversarial\ndemonstrations. We propose advCoU, an extended Chain of Utterances-based (CoU)\nprompting strategy by incorporating carefully crafted malicious demonstrations\nfor trustworthiness attack. Our extensive experiments encompass recent and\nrepresentative series of open-source LLMs, including Vicuna, MPT, Falcon,\nMistral, and Llama 2. The empirical outcomes underscore the efficacy of our\nattack strategy across diverse aspects. More interestingly, our result analysis\nreveals that models with superior performance in general NLP tasks do not\nalways have greater trustworthiness; in fact, larger models can be more\nvulnerable to attacks. Additionally, models that have undergone instruction\ntuning, focusing on instruction following, tend to be more susceptible,\nalthough fine-tuning LLMs for safety alignment proves effective in mitigating\nadversarial trustworthiness attacks.",
+        "authors": "Lingbo Mo, Boshi Wang, Muhao Chen, Huan Sun",
+        "published": "2023-11-15",
+        "updated": "2024-04-02",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.11595v3",
+        "title": "Examining Inter-Consistency of Large Language Models Collaboration: An In-depth Analysis via Debate",
+        "abstract": "Large Language Models (LLMs) have shown impressive capabilities in various\napplications, but they still face various inconsistency issues. Existing works\nprimarily focus on the inconsistency issues within a single LLM, while we\ncomplementarily explore the inter-consistency among multiple LLMs for\ncollaboration. To examine whether LLMs can collaborate effectively to achieve a\nconsensus for a shared goal, we focus on commonsense reasoning, and introduce a\nformal debate framework (FORD) to conduct a three-stage debate among LLMs with\nreal-world scenarios alignment: fair debate, mismatched debate, and roundtable\ndebate. Through extensive experiments on various datasets, LLMs can effectively\ncollaborate to reach a consensus despite noticeable inter-inconsistencies, but\nimbalances in their abilities can lead to domination by superior LLMs.\nLeveraging a more advanced LLM like GPT-4 as an authoritative judge can boost\ncollaboration performance. Our work contributes to understanding the\ninter-consistency among LLMs and lays the foundation for developing future\ncollaboration methods. Codes and data are available at\nhttps://github.com/Waste-Wood/FORD",
+        "authors": "Kai Xiong, Xiao Ding, Yixin Cao, Ting Liu, Bing Qin",
+        "published": "2023-05-19",
+        "updated": "2023-10-18",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.11406v2",
+        "title": "Don't Go To Extremes: Revealing the Excessive Sensitivity and Calibration Limitations of LLMs in Implicit Hate Speech Detection",
+        "abstract": "The fairness and trustworthiness of Large Language Models (LLMs) are\nreceiving increasing attention. Implicit hate speech, which employs indirect\nlanguage to convey hateful intentions, occupies a significant portion of\npractice. However, the extent to which LLMs effectively address this issue\nremains insufficiently examined. This paper delves into the capability of LLMs\nto detect implicit hate speech (Classification Task) and express confidence in\ntheir responses (Calibration Task). Our evaluation meticulously considers\nvarious prompt patterns and mainstream uncertainty estimation methods. Our\nfindings highlight that LLMs exhibit two extremes: (1) LLMs display excessive\nsensitivity towards groups or topics that may cause fairness issues, resulting\nin misclassifying benign statements as hate speech. (2) LLMs' confidence scores\nfor each method excessively concentrate on a fixed range, remaining unchanged\nregardless of the dataset's complexity. Consequently, the calibration\nperformance is heavily reliant on primary classification accuracy. These\ndiscoveries unveil new limitations of LLMs, underscoring the need for caution\nwhen optimizing models to ensure they do not veer towards extremes. This serves\nas a reminder to carefully consider sensitivity and confidence in the pursuit\nof model fairness.",
+        "authors": "Min Zhang, Jianfeng He, Taoran Ji, Chang-Tien Lu",
+        "published": "2024-02-18",
+        "updated": "2024-02-26",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.02294v1",
+        "title": "LLMs grasp morality in concept",
+        "abstract": "Work in AI ethics and fairness has made much progress in regulating LLMs to\nreflect certain values, such as fairness, truth, and diversity. However, it has\ntaken the problem of how LLMs might 'mean' anything at all for granted. Without\naddressing this, it is not clear what imbuing LLMs with such values even means.\nIn response, we provide a general theory of meaning that extends beyond humans.\nWe use this theory to explicate the precise nature of LLMs as meaning-agents.\nWe suggest that the LLM, by virtue of its position as a meaning-agent, already\ngrasps the constructions of human society (e.g. morality, gender, and race) in\nconcept. Consequently, under certain ethical frameworks, currently popular\nmethods for model alignment are limited at best and counterproductive at worst.\nMoreover, unaligned models may help us better develop our moral and social\nphilosophy.",
+        "authors": "Mark Pock, Andre Ye, Jared Moore",
+        "published": "2023-11-04",
+        "updated": "2023-11-04",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.18580v1",
+        "title": "FFT: Towards Harmlessness Evaluation and Analysis for LLMs with Factuality, Fairness, Toxicity",
+        "abstract": "The widespread of generative artificial intelligence has heightened concerns\nabout the potential harms posed by AI-generated texts, primarily stemming from\nfactoid, unfair, and toxic content. Previous researchers have invested much\neffort in assessing the harmlessness of generative language models. However,\nexisting benchmarks are struggling in the era of large language models (LLMs),\ndue to the stronger language generation and instruction following capabilities,\nas well as wider applications. In this paper, we propose FFT, a new benchmark\nwith 2116 elaborated-designed instances, for LLM harmlessness evaluation with\nfactuality, fairness, and toxicity. To investigate the potential harms of LLMs,\nwe evaluate 9 representative LLMs covering various parameter scales, training\nstages, and creators. Experiments show that the harmlessness of LLMs is still\nunder-satisfactory, and extensive analysis derives some insightful findings\nthat could inspire future research for harmless LLM research.",
+        "authors": "Shiyao Cui, Zhenyu Zhang, Yilong Chen, Wenyuan Zhang, Tianyun Liu, Siqi Wang, Tingwen Liu",
+        "published": "2023-11-30",
+        "updated": "2023-11-30",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.CR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.13840v1",
+        "title": "Whose Side Are You On? Investigating the Political Stance of Large Language Models",
+        "abstract": "Large Language Models (LLMs) have gained significant popularity for their\napplication in various everyday tasks such as text generation, summarization,\nand information retrieval. As the widespread adoption of LLMs continues to\nsurge, it becomes increasingly crucial to ensure that these models yield\nresponses that are politically impartial, with the aim of preventing\ninformation bubbles, upholding fairness in representation, and mitigating\nconfirmation bias. In this paper, we propose a quantitative framework and\npipeline designed to systematically investigate the political orientation of\nLLMs. Our investigation delves into the political alignment of LLMs across a\nspectrum of eight polarizing topics, spanning from abortion to LGBTQ issues.\nAcross topics, the results indicate that LLMs exhibit a tendency to provide\nresponses that closely align with liberal or left-leaning perspectives rather\nthan conservative or right-leaning ones when user queries include details\npertaining to occupation, race, or political affiliation. The findings\npresented in this study not only reaffirm earlier observations regarding the\nleft-leaning characteristics of LLMs but also surface particular attributes,\nsuch as occupation, that are particularly susceptible to such inclinations even\nwhen directly steered towards conservatism. As a recommendation to avoid these\nmodels providing politicised responses, users should be mindful when crafting\nqueries, and exercise caution in selecting neutral prompt language.",
+        "authors": "Pagnarasmey Pit, Xingjun Ma, Mike Conway, Qingyu Chen, James Bailey, Henry Pit, Putrasmey Keo, Watey Diep, Yu-Gang Jiang",
+        "published": "2024-03-15",
+        "updated": "2024-03-15",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.SI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2312.14769v3",
+        "title": "Large Language Model (LLM) Bias Index -- LLMBI",
+        "abstract": "The Large Language Model Bias Index (LLMBI) is a pioneering approach designed\nto quantify and address biases inherent in large language models (LLMs), such\nas GPT-4. We recognise the increasing prevalence and impact of LLMs across\ndiverse sectors. This research introduces a novel metric, LLMBI, to\nsystematically measure and mitigate biases potentially skewing model responses.\nWe formulated LLMBI using a composite scoring system incorporating multiple\ndimensions of bias, including but not limited to age, gender, and racial\nbiases. To operationalise this metric, we engaged in a multi-step process\ninvolving collecting and annotating LLM responses, applying sophisticated\nNatural Language Processing (NLP) techniques for bias detection, and computing\nthe LLMBI score through a specially crafted mathematical formula. The formula\nintegrates weighted averages of various bias dimensions, a penalty for dataset\ndiversity deficiencies, and a correction for sentiment biases. Our empirical\nanalysis, conducted using responses from OpenAI's API, employs advanced\nsentiment analysis as a representative method for bias detection. The research\nreveals LLMs, whilst demonstrating impressive capabilities in text generation,\nexhibit varying degrees of bias across different dimensions. LLMBI provides a\nquantifiable measure to compare biases across models and over time, offering a\nvital tool for systems engineers, researchers and regulators in enhancing the\nfairness and reliability of LLMs. It highlights the potential of LLMs in\nmimicking unbiased human-like responses. Additionally, it underscores the\nnecessity of continuously monitoring and recalibrating such models to align\nwith evolving societal norms and ethical standards.",
+        "authors": "Abiodun Finbarrs Oketunji, Muhammad Anas, Deepthi Saina",
+        "published": "2023-12-22",
+        "updated": "2023-12-29",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.CY",
+            "cs.LG",
+            "I.2.7"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2312.15398v1",
+        "title": "Fairness-Aware Structured Pruning in Transformers",
+        "abstract": "The increasing size of large language models (LLMs) has introduced challenges\nin their training and inference. Removing model components is perceived as a\nsolution to tackle the large model sizes, however, existing pruning methods\nsolely focus on performance, without considering an essential aspect for the\nresponsible use of LLMs: model fairness. It is crucial to address the fairness\nof LLMs towards diverse groups, such as women, Black people, LGBTQ+, Jewish\ncommunities, among others, as they are being deployed and available to a wide\naudience. In this work, first, we investigate how attention heads impact\nfairness and performance in pre-trained transformer-based language models. We\nthen propose a novel method to prune the attention heads that negatively impact\nfairness while retaining the heads critical for performance, i.e. language\nmodeling capabilities. Our approach is practical in terms of time and\nresources, as it does not require fine-tuning the final pruned, and fairer,\nmodel. Our findings demonstrate a reduction in gender bias by 19%, 19.5%,\n39.5%, 34.7%, 23%, and 8% for DistilGPT-2, GPT-2, GPT-Neo of two different\nsizes, GPT-J, and Llama 2 models, respectively, in comparison to the biased\nmodel, with only a slight decrease in performance.",
+        "authors": "Abdelrahman Zayed, Goncalo Mordido, Samira Shabanian, Ioana Baldini, Sarath Chandar",
+        "published": "2023-12-24",
+        "updated": "2023-12-24",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.CY",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.01964v1",
+        "title": "Don't Make Your LLM an Evaluation Benchmark Cheater",
+        "abstract": "Large language models~(LLMs) have greatly advanced the frontiers of\nartificial intelligence, attaining remarkable improvement in model capacity. To\nassess the model performance, a typical approach is to construct evaluation\nbenchmarks for measuring the ability level of LLMs in different aspects.\nDespite that a number of high-quality benchmarks have been released, the\nconcerns about the appropriate use of these benchmarks and the fair comparison\nof different models are increasingly growing. Considering these concerns, in\nthis paper, we discuss the potential risk and impact of inappropriately using\nevaluation benchmarks and misleadingly interpreting the evaluation results.\nSpecially, we focus on a special issue that would lead to inappropriate\nevaluation, \\ie \\emph{benchmark leakage}, referring that the data related to\nevaluation sets is occasionally used for model training. This phenomenon now\nbecomes more common since pre-training data is often prepared ahead of model\ntest. We conduct extensive experiments to study the effect of benchmark\nleverage, and find that it can dramatically boost the evaluation results, which\nwould finally lead to an unreliable assessment of model performance. To improve\nthe use of existing evaluation benchmarks, we finally present several\nguidelines for both LLM developers and benchmark maintainers. We hope this work\ncan draw attention to appropriate training and evaluation of LLMs.",
+        "authors": "Kun Zhou, Yutao Zhu, Zhipeng Chen, Wentong Chen, Wayne Xin Zhao, Xu Chen, Yankai Lin, Ji-Rong Wen, Jiawei Han",
+        "published": "2023-11-03",
+        "updated": "2023-11-03",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2312.14804v1",
+        "title": "Use large language models to promote equity",
+        "abstract": "Advances in large language models (LLMs) have driven an explosion of interest\nabout their societal impacts. Much of the discourse around how they will impact\nsocial equity has been cautionary or negative, focusing on questions like \"how\nmight LLMs be biased and how would we mitigate those biases?\" This is a vital\ndiscussion: the ways in which AI generally, and LLMs specifically, can entrench\nbiases have been well-documented. But equally vital, and much less discussed,\nis the more opportunity-focused counterpoint: \"what promising applications do\nLLMs enable that could promote equity?\" If LLMs are to enable a more equitable\nworld, it is not enough just to play defense against their biases and failure\nmodes. We must also go on offense, applying them positively to equity-enhancing\nuse cases to increase opportunities for underserved groups and reduce societal\ndiscrimination. There are many choices which determine the impact of AI, and a\nfundamental choice very early in the pipeline is the problems we choose to\napply it to. If we focus only later in the pipeline -- making LLMs marginally\nmore fair as they facilitate use cases which intrinsically entrench power -- we\nwill miss an important opportunity to guide them to equitable impacts. Here, we\nhighlight the emerging potential of LLMs to promote equity by presenting four\nnewly possible, promising research directions, while keeping risks and\ncautionary points in clear view.",
+        "authors": "Emma Pierson, Divya Shanmugam, Rajiv Movva, Jon Kleinberg, Monica Agrawal, Mark Dredze, Kadija Ferryman, Judy Wawira Gichoya, Dan Jurafsky, Pang Wei Koh, Karen Levy, Sendhil Mullainathan, Ziad Obermeyer, Harini Suresh, Keyon Vafa",
+        "published": "2023-12-22",
+        "updated": "2023-12-22",
+        "primary_cat": "cs.CY",
+        "cats": [
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.07981v1",
+        "title": "Manipulating Large Language Models to Increase Product Visibility",
+        "abstract": "Large language models (LLMs) are increasingly being integrated into search\nengines to provide natural language responses tailored to user queries.\nCustomers and end-users are also becoming more dependent on these models for\nquick and easy purchase decisions. In this work, we investigate whether\nrecommendations from LLMs can be manipulated to enhance a product's visibility.\nWe demonstrate that adding a strategic text sequence (STS) -- a carefully\ncrafted message -- to a product's information page can significantly increase\nits likelihood of being listed as the LLM's top recommendation. To understand\nthe impact of STS, we use a catalog of fictitious coffee machines and analyze\nits effect on two target products: one that seldom appears in the LLM's\nrecommendations and another that usually ranks second. We observe that the\nstrategic text sequence significantly enhances the visibility of both products\nby increasing their chances of appearing as the top recommendation. This\nability to manipulate LLM-generated search responses provides vendors with a\nconsiderable competitive advantage and has the potential to disrupt fair market\ncompetition. Just as search engine optimization (SEO) revolutionized how\nwebpages are customized to rank higher in search engine results, influencing\nLLM recommendations could profoundly impact content optimization for AI-driven\nsearch services. Code for our experiments is available at\nhttps://github.com/aounon/llm-rank-optimizer.",
+        "authors": "Aounon Kumar, Himabindu Lakkaraju",
+        "published": "2024-04-11",
+        "updated": "2024-04-11",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR",
+            "cs.AI",
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.02680v1",
+        "title": "Large Language Models are Geographically Biased",
+        "abstract": "Large Language Models (LLMs) inherently carry the biases contained in their\ntraining corpora, which can lead to the perpetuation of societal harm. As the\nimpact of these foundation models grows, understanding and evaluating their\nbiases becomes crucial to achieving fairness and accuracy. We propose to study\nwhat LLMs know about the world we live in through the lens of geography. This\napproach is particularly powerful as there is ground truth for the numerous\naspects of human life that are meaningfully projected onto geographic space\nsuch as culture, race, language, politics, and religion. We show various\nproblematic geographic biases, which we define as systemic errors in geospatial\npredictions. Initially, we demonstrate that LLMs are capable of making accurate\nzero-shot geospatial predictions in the form of ratings that show strong\nmonotonic correlation with ground truth (Spearman's $\\rho$ of up to 0.89). We\nthen show that LLMs exhibit common biases across a range of objective and\nsubjective topics. In particular, LLMs are clearly biased against locations\nwith lower socioeconomic conditions (e.g. most of Africa) on a variety of\nsensitive subjective topics such as attractiveness, morality, and intelligence\n(Spearman's $\\rho$ of up to 0.70). Finally, we introduce a bias score to\nquantify this and find that there is significant variation in the magnitude of\nbias across existing LLMs.",
+        "authors": "Rohin Manvi, Samar Khanna, Marshall Burke, David Lobell, Stefano Ermon",
+        "published": "2024-02-05",
+        "updated": "2024-02-05",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.CY",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.15215v1",
+        "title": "Item-side Fairness of Large Language Model-based Recommendation System",
+        "abstract": "Recommendation systems for Web content distribution intricately connect to\nthe information access and exposure opportunities for vulnerable populations.\nThe emergence of Large Language Models-based Recommendation System (LRS) may\nintroduce additional societal challenges to recommendation systems due to the\ninherent biases in Large Language Models (LLMs). From the perspective of\nitem-side fairness, there remains a lack of comprehensive investigation into\nthe item-side fairness of LRS given the unique characteristics of LRS compared\nto conventional recommendation systems. To bridge this gap, this study examines\nthe property of LRS with respect to item-side fairness and reveals the\ninfluencing factors of both historical users' interactions and inherent\nsemantic biases of LLMs, shedding light on the need to extend conventional\nitem-side fairness methods for LRS. Towards this goal, we develop a concise and\neffective framework called IFairLRS to enhance the item-side fairness of an\nLRS. IFairLRS covers the main stages of building an LRS with specifically\nadapted strategies to calibrate the recommendations of LRS. We utilize IFairLRS\nto fine-tune LLaMA, a representative LLM, on \\textit{MovieLens} and\n\\textit{Steam} datasets, and observe significant item-side fairness\nimprovements. The code can be found in\nhttps://github.com/JiangM-C/IFairLRS.git.",
+        "authors": "Meng Jiang, Keqin Bao, Jizhi Zhang, Wenjie Wang, Zhengyi Yang, Fuli Feng, Xiangnan He",
+        "published": "2024-02-23",
+        "updated": "2024-02-23",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2405.02219v1",
+        "title": "FairEvalLLM. A Comprehensive Framework for Benchmarking Fairness in Large Language Model Recommender Systems",
+        "abstract": "This paper presents a framework for evaluating fairness in recommender\nsystems powered by Large Language Models (RecLLMs), addressing the need for a\nunified approach that spans various fairness dimensions including sensitivity\nto user attributes, intrinsic fairness, and discussions of fairness based on\nunderlying benefits. In addition, our framework introduces counterfactual\nevaluations and integrates diverse user group considerations to enhance the\ndiscourse on fairness evaluation for RecLLMs.\n  Our key contributions include the development of a robust framework for\nfairness evaluation in LLM-based recommendations and a structured method to\ncreate \\textit{informative user profiles} from demographic data, historical\nuser preferences, and recent interactions. We argue that the latter is\nessential for enhancing personalization in such systems, especially in\ntemporal-driven scenarios. We demonstrate the utility of our framework through\npractical applications on two datasets, LastFM-1K and ML-1M. We conduct\nexperiments on a subsample of 80 users from each dataset, testing and assessing\nthe effectiveness of various prompt construction scenarios and in-context\nlearning, comprising more than 50 scenarios. This results in more than 4000\nrecommendations (80 * 50 = 4000). Our study reveals that while there are no\nsignificant unfairness issues in scenarios involving sensitive attributes, some\nconcerns remain. However, in terms of intrinsic fairness, which does not\ninvolve direct sensitivity, unfairness across demographic groups remains\nsignificant. The code and data used for this paper are available at:\n\\url{https://shorturl.at/awBFM}.",
+        "authors": "Yashar Deldjoo",
+        "published": "2024-05-03",
+        "updated": "2024-05-03",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.16343v2",
+        "title": "Evaluating, Understanding, and Improving Constrained Text Generation for Large Language Models",
+        "abstract": "Advancements in natural language generation (NLG) and large language models\n(LLMs) have led to proficient text generation in various tasks. However,\nintegrating intricate constraints into neural text generation, due to LLMs'\nopacity, remains challenging. This study investigates constrained text\ngeneration for LLMs, where predefined constraints are applied during LLM's\ngeneration process. Our research mainly focuses on mainstream open-source LLMs,\ncategorizing constraints into lexical, structural, and relation-based types. We\nalso present various benchmarks to facilitate fair evaluation. The study\naddresses some key research questions, including evaluating, understanding and\nimproving constrained text generation for LLMs. Results illuminate LLMs'\ncapacity and deficiency to incorporate constraints and provide insights for\nfuture developments in constrained text generation. Codes and datasets will be\nreleased upon acceptance.",
+        "authors": "Xiang Chen, Xiaojun Wan",
+        "published": "2023-10-25",
+        "updated": "2024-03-21",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.09606v1",
+        "title": "Large Language Models and Causal Inference in Collaboration: A Comprehensive Survey",
+        "abstract": "Causal inference has shown potential in enhancing the predictive accuracy,\nfairness, robustness, and explainability of Natural Language Processing (NLP)\nmodels by capturing causal relationships among variables. The emergence of\ngenerative Large Language Models (LLMs) has significantly impacted various NLP\ndomains, particularly through their advanced reasoning capabilities. This\nsurvey focuses on evaluating and improving LLMs from a causal view in the\nfollowing areas: understanding and improving the LLMs' reasoning capacity,\naddressing fairness and safety issues in LLMs, complementing LLMs with\nexplanations, and handling multimodality. Meanwhile, LLMs' strong reasoning\ncapacities can in turn contribute to the field of causal inference by aiding\ncausal relationship discovery and causal effect estimations. This review\nexplores the interplay between causal inference frameworks and LLMs from both\nperspectives, emphasizing their collective potential to further the development\nof more advanced and equitable artificial intelligence systems.",
+        "authors": "Xiaoyu Liu, Paiheng Xu, Junda Wu, Jiaxin Yuan, Yifan Yang, Yuhang Zhou, Fuxiao Liu, Tianrui Guan, Haoliang Wang, Tong Yu, Julian McAuley, Wei Ai, Furong Huang",
+        "published": "2024-03-14",
+        "updated": "2024-03-14",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.18140v1",
+        "title": "ROBBIE: Robust Bias Evaluation of Large Generative Language Models",
+        "abstract": "As generative large language models (LLMs) grow more performant and\nprevalent, we must develop comprehensive enough tools to measure and improve\ntheir fairness. Different prompt-based datasets can be used to measure social\nbias across multiple text domains and demographic axes, meaning that testing\nLLMs on more datasets can potentially help us characterize their biases more\nfully, and better ensure equal and equitable treatment of marginalized\ndemographic groups. In this work, our focus is two-fold:\n  (1) Benchmarking: a comparison of 6 different prompt-based bias and toxicity\nmetrics across 12 demographic axes and 5 families of generative LLMs. Out of\nthose 6 metrics, AdvPromptSet and HolisticBiasR are novel datasets proposed in\nthe paper. The comparison of those benchmarks gives us insights about the bias\nand toxicity of the compared models. Therefore, we explore the frequency of\ndemographic terms in common LLM pre-training corpora and how this may relate to\nmodel biases.\n  (2) Mitigation: we conduct a comprehensive study of how well 3 bias/toxicity\nmitigation techniques perform across our suite of measurements. ROBBIE aims to\nprovide insights for practitioners while deploying a model, emphasizing the\nneed to not only measure potential harms, but also understand how they arise by\ncharacterizing the data, mitigate harms once found, and balance any trade-offs.\nWe open-source our analysis code in hopes of encouraging broader measurements\nof bias in future LLMs.",
+        "authors": "David Esiobu, Xiaoqing Tan, Saghar Hosseini, Megan Ung, Yuchen Zhang, Jude Fernandes, Jane Dwivedi-Yu, Eleonora Presani, Adina Williams, Eric Michael Smith",
+        "published": "2023-11-29",
+        "updated": "2023-11-29",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.06899v4",
+        "title": "Flames: Benchmarking Value Alignment of LLMs in Chinese",
+        "abstract": "The widespread adoption of large language models (LLMs) across various\nregions underscores the urgent need to evaluate their alignment with human\nvalues. Current benchmarks, however, fall short of effectively uncovering\nsafety vulnerabilities in LLMs. Despite numerous models achieving high scores\nand 'topping the chart' in these evaluations, there is still a significant gap\nin LLMs' deeper alignment with human values and achieving genuine harmlessness.\nTo this end, this paper proposes a value alignment benchmark named Flames,\nwhich encompasses both common harmlessness principles and a unique morality\ndimension that integrates specific Chinese values such as harmony. Accordingly,\nwe carefully design adversarial prompts that incorporate complex scenarios and\njailbreaking methods, mostly with implicit malice. By prompting 17 mainstream\nLLMs, we obtain model responses and rigorously annotate them for detailed\nevaluation. Our findings indicate that all the evaluated LLMs demonstrate\nrelatively poor performance on Flames, particularly in the safety and fairness\ndimensions. We also develop a lightweight specified scorer capable of scoring\nLLMs across multiple dimensions to efficiently evaluate new models on the\nbenchmark. The complexity of Flames has far exceeded existing benchmarks,\nsetting a new challenge for contemporary LLMs and highlighting the need for\nfurther alignment of LLMs. Our benchmark is publicly available at\nhttps://github.com/AIFlames/Flames.",
+        "authors": "Kexin Huang, Xiangyang Liu, Qianyu Guo, Tianxiang Sun, Jiawei Sun, Yaru Wang, Zeyang Zhou, Yixu Wang, Yan Teng, Xipeng Qiu, Yingchun Wang, Dahua Lin",
+        "published": "2023-11-12",
+        "updated": "2024-04-15",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.14473v1",
+        "title": "The Ethics of ChatGPT in Medicine and Healthcare: A Systematic Review on Large Language Models (LLMs)",
+        "abstract": "With the introduction of ChatGPT, Large Language Models (LLMs) have received\nenormous attention in healthcare. Despite their potential benefits, researchers\nhave underscored various ethical implications. While individual instances have\ndrawn much attention, the debate lacks a systematic overview of practical\napplications currently researched and ethical issues connected to them. Against\nthis background, this work aims to map the ethical landscape surrounding the\ncurrent stage of deployment of LLMs in medicine and healthcare. Electronic\ndatabases and preprint servers were queried using a comprehensive search\nstrategy. Studies were screened and extracted following a modified rapid review\napproach. Methodological quality was assessed using a hybrid approach. For 53\nrecords, a meta-aggregative synthesis was performed. Four fields of\napplications emerged and testify to a vivid exploration phase. Advantages of\nusing LLMs are attributed to their capacity in data analysis, personalized\ninformation provisioning, support in decision-making, mitigating information\nloss and enhancing information accessibility. However, we also identifies\nrecurrent ethical concerns connected to fairness, bias, non-maleficence,\ntransparency, and privacy. A distinctive concern is the tendency to produce\nharmful misinformation or convincingly but inaccurate content. A recurrent plea\nfor ethical guidance and human oversight is evident. Given the variety of use\ncases, it is suggested that the ethical guidance debate be reframed to focus on\ndefining what constitutes acceptable human oversight across the spectrum of\napplications. This involves considering diverse settings, varying potentials\nfor harm, and different acceptable thresholds for performance and certainty in\nhealthcare. In addition, a critical inquiry is necessary to determine the\nextent to which the current experimental use of LLMs is necessary and\njustified.",
+        "authors": "Joschka Haltaufderheide, Robert Ranisch",
+        "published": "2024-03-21",
+        "updated": "2024-03-21",
+        "primary_cat": "cs.CY",
+        "cats": [
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.01349v1",
+        "title": "Fairness in Large Language Models: A Taxonomic Survey",
+        "abstract": "Large Language Models (LLMs) have demonstrated remarkable success across\nvarious domains. However, despite their promising performance in numerous\nreal-world applications, most of these algorithms lack fairness considerations.\nConsequently, they may lead to discriminatory outcomes against certain\ncommunities, particularly marginalized populations, prompting extensive study\nin fair LLMs. On the other hand, fairness in LLMs, in contrast to fairness in\ntraditional machine learning, entails exclusive backgrounds, taxonomies, and\nfulfillment techniques. To this end, this survey presents a comprehensive\noverview of recent advances in the existing literature concerning fair LLMs.\nSpecifically, a brief introduction to LLMs is provided, followed by an analysis\nof factors contributing to bias in LLMs. Additionally, the concept of fairness\nin LLMs is discussed categorically, summarizing metrics for evaluating bias in\nLLMs and existing algorithms for promoting fairness. Furthermore, resources for\nevaluating bias in LLMs, including toolkits and datasets, are summarized.\nFinally, existing research challenges and open questions are discussed.",
+        "authors": "Zhibo Chu, Zichong Wang, Wenbin Zhang",
+        "published": "2024-03-31",
+        "updated": "2024-03-31",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.03192v1",
+        "title": "Do Large Language Models Rank Fairly? An Empirical Study on the Fairness of LLMs as Rankers",
+        "abstract": "The integration of Large Language Models (LLMs) in information retrieval has\nraised a critical reevaluation of fairness in the text-ranking models. LLMs,\nsuch as GPT models and Llama2, have shown effectiveness in natural language\nunderstanding tasks, and prior works (e.g., RankGPT) have also demonstrated\nthat the LLMs exhibit better performance than the traditional ranking models in\nthe ranking task. However, their fairness remains largely unexplored. This\npaper presents an empirical study evaluating these LLMs using the TREC Fair\nRanking dataset, focusing on the representation of binary protected attributes\nsuch as gender and geographic location, which are historically underrepresented\nin search outcomes. Our analysis delves into how these LLMs handle queries and\ndocuments related to these attributes, aiming to uncover biases in their\nranking algorithms. We assess fairness from both user and content perspectives,\ncontributing an empirical benchmark for evaluating LLMs as the fair ranker.",
+        "authors": "Yuan Wang, Xuyang Wu, Hsin-Tai Wu, Zhiqiang Tao, Yi Fang",
+        "published": "2024-04-04",
+        "updated": "2024-04-04",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR",
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.10199v3",
+        "title": "CULTURE-GEN: Revealing Global Cultural Perception in Language Models through Natural Language Prompting",
+        "abstract": "As the utilization of large language models (LLMs) has proliferated\nworldwide, it is crucial for them to have adequate knowledge and fair\nrepresentation for diverse global cultures. In this work, we uncover culture\nperceptions of three SOTA models on 110 countries and regions on 8\nculture-related topics through culture-conditioned generations, and extract\nsymbols from these generations that are associated to each culture by the LLM.\nWe discover that culture-conditioned generation consist of linguistic \"markers\"\nthat distinguish marginalized cultures apart from default cultures. We also\ndiscover that LLMs have an uneven degree of diversity in the culture symbols,\nand that cultures from different geographic regions have different presence in\nLLMs' culture-agnostic generation. Our findings promote further research in\nstudying the knowledge and fairness of global culture perception in LLMs. Code\nand Data can be found in: https://github.com/huihanlhh/Culture-Gen/",
+        "authors": "Huihan Li, Liwei Jiang, Jena D. Huang, Hyunwoo Kim, Sebastin Santy, Taylor Sorensen, Bill Yuchen Lin, Nouha Dziri, Xiang Ren, Yejin Choi",
+        "published": "2024-04-16",
+        "updated": "2024-04-26",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.10567v3",
+        "title": "InSaAF: Incorporating Safety through Accuracy and Fairness | Are LLMs ready for the Indian Legal Domain?",
+        "abstract": "Recent advancements in language technology and Artificial Intelligence have\nresulted in numerous Language Models being proposed to perform various tasks in\nthe legal domain ranging from predicting judgments to generating summaries.\nDespite their immense potential, these models have been proven to learn and\nexhibit societal biases and make unfair predictions. In this study, we explore\nthe ability of Large Language Models (LLMs) to perform legal tasks in the\nIndian landscape when social factors are involved. We present a novel metric,\n$\\beta$-weighted $\\textit{Legal Safety Score ($LSS_{\\beta}$)}$, which\nencapsulates both the fairness and accuracy aspects of the LLM. We assess LLMs'\nsafety by considering its performance in the $\\textit{Binary Statutory\nReasoning}$ task and its fairness exhibition with respect to various axes of\ndisparities in the Indian society. Task performance and fairness scores of\nLLaMA and LLaMA--2 models indicate that the proposed $LSS_{\\beta}$ metric can\neffectively determine the readiness of a model for safe usage in the legal\nsector. We also propose finetuning pipelines, utilising specialised legal\ndatasets, as a potential method to mitigate bias and improve model safety. The\nfinetuning procedures on LLaMA and LLaMA--2 models increase the $LSS_{\\beta}$,\nimproving their usability in the Indian legal domain. Our code is publicly\nreleased.",
+        "authors": "Yogesh Tripathi, Raghav Donakanti, Sahil Girhepuje, Ishan Kavathekar, Bhaskara Hanuma Vedula, Gokul S Krishnan, Shreya Goyal, Anmol Goel, Balaraman Ravindran, Ponnurangam Kumaraguru",
+        "published": "2024-02-16",
+        "updated": "2024-02-21",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2206.13757v1",
+        "title": "Flexible text generation for counterfactual fairness probing",
+        "abstract": "A common approach for testing fairness issues in text-based classifiers is\nthrough the use of counterfactuals: does the classifier output change if a\nsensitive attribute in the input is changed? Existing counterfactual generation\nmethods typically rely on wordlists or templates, producing simple\ncounterfactuals that don't take into account grammar, context, or subtle\nsensitive attribute references, and could miss issues that the wordlist\ncreators had not considered. In this paper, we introduce a task for generating\ncounterfactuals that overcomes these shortcomings, and demonstrate how large\nlanguage models (LLMs) can be leveraged to make progress on this task. We show\nthat this LLM-based method can produce complex counterfactuals that existing\nmethods cannot, comparing the performance of various counterfactual generation\nmethods on the Civil Comments dataset and showing their value in evaluating a\ntoxicity classifier.",
+        "authors": "Zee Fryer, Vera Axelrod, Ben Packer, Alex Beutel, Jilin Chen, Kellie Webster",
+        "published": "2022-06-28",
+        "updated": "2022-06-28",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.00884v2",
+        "title": "Text classification of column headers with a controlled vocabulary: leveraging LLMs for metadata enrichment",
+        "abstract": "Traditional dataset retrieval systems index on metadata information rather\nthan on the data values. Thus relying primarily on manual annotations and\nhigh-quality metadata, processes known to be labour-intensive and challenging\nto automate. We propose a method to support metadata enrichment with topic\nannotations of column headers using three Large Language Models (LLMs):\nChatGPT-3.5, GoogleBard and GoogleGemini. We investigate the LLMs ability to\nclassify column headers based on domain-specific topics from a controlled\nvocabulary. We evaluate our approach by assessing the internal consistency of\nthe LLMs, the inter-machine alignment, and the human-machine agreement for the\ntopic classification task. Additionally, we investigate the impact of\ncontextual information (i.e. dataset description) on the classification\noutcomes. Our results suggest that ChatGPT and GoogleGemini outperform\nGoogleBard for internal consistency as well as LLM-human-alignment.\nInterestingly, we found that context had no impact on the LLMs performances.\nThis work proposes a novel approach that leverages LLMs for text classification\nusing a controlled topic vocabulary, which has the potential to facilitate\nautomated metadata enrichment, thereby enhancing dataset retrieval and the\nFindability, Accessibility, Interoperability and Reusability (FAIR) of research\ndata on the Web.",
+        "authors": "Margherita Martorana, Tobias Kuhn, Lise Stork, Jacco van Ossenbruggen",
+        "published": "2024-03-01",
+        "updated": "2024-03-05",
+        "primary_cat": "cs.DB",
+        "cats": [
+            "cs.DB",
+            "cs.AI",
+            "cs.IR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.00306v1",
+        "title": "Probing Explicit and Implicit Gender Bias through LLM Conditional Text Generation",
+        "abstract": "Large Language Models (LLMs) can generate biased and toxic responses. Yet\nmost prior work on LLM gender bias evaluation requires predefined\ngender-related phrases or gender stereotypes, which are challenging to be\ncomprehensively collected and are limited to explicit bias evaluation. In\naddition, we believe that instances devoid of gender-related language or\nexplicit stereotypes in inputs can still induce gender bias in LLMs. Thus, in\nthis work, we propose a conditional text generation mechanism without the need\nfor predefined gender phrases and stereotypes. This approach employs three\ntypes of inputs generated through three distinct strategies to probe LLMs,\naiming to show evidence of explicit and implicit gender biases in LLMs. We also\nutilize explicit and implicit evaluation metrics to evaluate gender bias in\nLLMs under different strategies. Our experiments demonstrate that an increased\nmodel size does not consistently lead to enhanced fairness and all tested LLMs\nexhibit explicit and/or implicit gender bias, even when explicit gender\nstereotypes are absent in the inputs.",
+        "authors": "Xiangjue Dong, Yibo Wang, Philip S. Yu, James Caverlee",
+        "published": "2023-11-01",
+        "updated": "2023-11-01",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2309.11653v2",
+        "title": "\"It's a Fair Game\", or Is It? Examining How Users Navigate Disclosure Risks and Benefits When Using LLM-Based Conversational Agents",
+        "abstract": "The widespread use of Large Language Model (LLM)-based conversational agents\n(CAs), especially in high-stakes domains, raises many privacy concerns.\nBuilding ethical LLM-based CAs that respect user privacy requires an in-depth\nunderstanding of the privacy risks that concern users the most. However,\nexisting research, primarily model-centered, does not provide insight into\nusers' perspectives. To bridge this gap, we analyzed sensitive disclosures in\nreal-world ChatGPT conversations and conducted semi-structured interviews with\n19 LLM-based CA users. We found that users are constantly faced with trade-offs\nbetween privacy, utility, and convenience when using LLM-based CAs. However,\nusers' erroneous mental models and the dark patterns in system design limited\ntheir awareness and comprehension of the privacy risks. Additionally, the\nhuman-like interactions encouraged more sensitive disclosures, which\ncomplicated users' ability to navigate the trade-offs. We discuss practical\ndesign guidelines and the needs for paradigm shifts to protect the privacy of\nLLM-based CA users.",
+        "authors": "Zhiping Zhang, Michelle Jia, Hao-Ping Lee, Bingsheng Yao, Sauvik Das, Ada Lerner, Dakuo Wang, Tianshi Li",
+        "published": "2023-09-20",
+        "updated": "2024-04-02",
+        "primary_cat": "cs.HC",
+        "cats": [
+            "cs.HC",
+            "cs.AI",
+            "cs.CR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.05694v1",
+        "title": "A Survey of Large Language Models for Healthcare: from Data, Technology, and Applications to Accountability and Ethics",
+        "abstract": "The utilization of large language models (LLMs) in the Healthcare domain has\ngenerated both excitement and concern due to their ability to effectively\nrespond to freetext queries with certain professional knowledge. This survey\noutlines the capabilities of the currently developed LLMs for Healthcare and\nexplicates their development process, with the aim of providing an overview of\nthe development roadmap from traditional Pretrained Language Models (PLMs) to\nLLMs. Specifically, we first explore the potential of LLMs to enhance the\nefficiency and effectiveness of various Healthcare applications highlighting\nboth the strengths and limitations. Secondly, we conduct a comparison between\nthe previous PLMs and the latest LLMs, as well as comparing various LLMs with\neach other. Then we summarize related Healthcare training data, training\nmethods, optimization strategies, and usage. Finally, the unique concerns\nassociated with deploying LLMs in Healthcare settings are investigated,\nparticularly regarding fairness, accountability, transparency and ethics. Our\nsurvey provide a comprehensive investigation from perspectives of both computer\nscience and Healthcare specialty. Besides the discussion about Healthcare\nconcerns, we supports the computer science community by compiling a collection\nof open source resources, such as accessible datasets, the latest\nmethodologies, code implementations, and evaluation benchmarks in the Github.\nSummarily, we contend that a significant paradigm shift is underway,\ntransitioning from PLMs to LLMs. This shift encompasses a move from\ndiscriminative AI approaches to generative AI approaches, as well as a shift\nfrom model-centered methodologies to datacentered methodologies.",
+        "authors": "Kai He, Rui Mao, Qika Lin, Yucheng Ruan, Xiang Lan, Mengling Feng, Erik Cambria",
+        "published": "2023-10-09",
+        "updated": "2023-10-09",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.15007v1",
+        "title": "Did the Neurons Read your Book? Document-level Membership Inference for Large Language Models",
+        "abstract": "With large language models (LLMs) poised to become embedded in our daily\nlives, questions are starting to be raised about the dataset(s) they learned\nfrom. These questions range from potential bias or misinformation LLMs could\nretain from their training data to questions of copyright and fair use of\nhuman-generated text. However, while these questions emerge, developers of the\nrecent state-of-the-art LLMs become increasingly reluctant to disclose details\non their training corpus. We here introduce the task of document-level\nmembership inference for real-world LLMs, i.e. inferring whether the LLM has\nseen a given document during training or not. First, we propose a procedure for\nthe development and evaluation of document-level membership inference for LLMs\nby leveraging commonly used data sources for training and the model release\ndate. We then propose a practical, black-box method to predict document-level\nmembership and instantiate it on OpenLLaMA-7B with both books and academic\npapers. We show our methodology to perform very well, reaching an impressive\nAUC of 0.856 for books and 0.678 for papers. We then show our approach to\noutperform the sentence-level membership inference attacks used in the privacy\nliterature for the document-level membership task. We finally evaluate whether\nsmaller models might be less sensitive to document-level inference and show\nOpenLLaMA-3B to be approximately as sensitive as OpenLLaMA-7B to our approach.\nTaken together, our results show that accurate document-level membership can be\ninferred for LLMs, increasing the transparency of technology poised to change\nour lives.",
+        "authors": "Matthieu Meeus, Shubham Jain, Marek Rei, Yves-Alexandre de Montjoye",
+        "published": "2023-10-23",
+        "updated": "2023-10-23",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.CR",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.04892v2",
+        "title": "Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs",
+        "abstract": "Recent works have showcased the ability of LLMs to embody diverse personas in\ntheir responses, exemplified by prompts like 'You are Yoda. Explain the Theory\nof Relativity.' While this ability allows personalization of LLMs and enables\nhuman behavior simulation, its effect on LLMs' capabilities remains unclear. To\nfill this gap, we present the first extensive study of the unintended\nside-effects of persona assignment on the ability of LLMs to perform basic\nreasoning tasks. Our study covers 24 reasoning datasets, 4 LLMs, and 19 diverse\npersonas (e.g. an Asian person) spanning 5 socio-demographic groups. Our\nexperiments unveil that LLMs harbor deep rooted bias against various\nsocio-demographics underneath a veneer of fairness. While they overtly reject\nstereotypes when explicitly asked ('Are Black people less skilled at\nmathematics?'), they manifest stereotypical and erroneous presumptions when\nasked to answer questions while adopting a persona. These can be observed as\nabstentions in responses, e.g., 'As a Black person, I can't answer this\nquestion as it requires math knowledge', and generally result in a substantial\nperformance drop. Our experiments with ChatGPT-3.5 show that this bias is\nubiquitous - 80% of our personas demonstrate bias; it is significant - some\ndatasets show performance drops of 70%+; and can be especially harmful for\ncertain groups - some personas suffer statistically significant drops on 80%+\nof the datasets. Overall, all 4 LLMs exhibit this bias to varying extents, with\nGPT-4-Turbo showing the least but still a problematic amount of bias (evident\nin 42% of the personas). Further analysis shows that these persona-induced\nerrors can be hard-to-discern and hard-to-avoid. Our findings serve as a\ncautionary tale that the practice of assigning personas to LLMs - a trend on\nthe rise - can surface their deep-rooted biases and have unforeseeable and\ndetrimental side-effects.",
+        "authors": "Shashank Gupta, Vaishnavi Shrivastava, Ameet Deshpande, Ashwin Kalyan, Peter Clark, Ashish Sabharwal, Tushar Khot",
+        "published": "2023-11-08",
+        "updated": "2024-01-27",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2401.00625v2",
+        "title": "Beyond Efficiency: A Systematic Survey of Resource-Efficient Large Language Models",
+        "abstract": "The burgeoning field of Large Language Models (LLMs), exemplified by\nsophisticated models like OpenAI's ChatGPT, represents a significant\nadvancement in artificial intelligence. These models, however, bring forth\nsubstantial challenges in the high consumption of computational, memory,\nenergy, and financial resources, especially in environments with limited\nresource capabilities. This survey aims to systematically address these\nchallenges by reviewing a broad spectrum of techniques designed to enhance the\nresource efficiency of LLMs. We categorize methods based on their optimization\nfocus: computational, memory, energy, financial, and network resources and\ntheir applicability across various stages of an LLM's lifecycle, including\narchitecture design, pretraining, finetuning, and system design. Additionally,\nthe survey introduces a nuanced categorization of resource efficiency\ntechniques by their specific resource types, which uncovers the intricate\nrelationships and mappings between various resources and corresponding\noptimization techniques. A standardized set of evaluation metrics and datasets\nis also presented to facilitate consistent and fair comparisons across\ndifferent models and techniques. By offering a comprehensive overview of the\ncurrent sota and identifying open research avenues, this survey serves as a\nfoundational reference for researchers and practitioners, aiding them in\ndeveloping more sustainable and efficient LLMs in a rapidly evolving landscape.",
+        "authors": "Guangji Bai, Zheng Chai, Chen Ling, Shiyu Wang, Jiaying Lu, Nan Zhang, Tingwei Shi, Ziyang Yu, Mengdan Zhu, Yifei Zhang, Carl Yang, Yue Cheng, Liang Zhao",
+        "published": "2024-01-01",
+        "updated": "2024-01-04",
+        "primary_cat": "cs.LG",
+        "cats": [
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.04205v2",
+        "title": "Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselves",
+        "abstract": "Misunderstandings arise not only in interpersonal communication but also\nbetween humans and Large Language Models (LLMs). Such discrepancies can make\nLLMs interpret seemingly unambiguous questions in unexpected ways, yielding\nincorrect responses. While it is widely acknowledged that the quality of a\nprompt, such as a question, significantly impacts the quality of the response\nprovided by LLMs, a systematic method for crafting questions that LLMs can\nbetter comprehend is still underdeveloped. In this paper, we present a method\nnamed `Rephrase and Respond' (RaR), which allows LLMs to rephrase and expand\nquestions posed by humans and provide responses in a single prompt. This\napproach serves as a simple yet effective prompting method for improving\nperformance. We also introduce a two-step variant of RaR, where a rephrasing\nLLM first rephrases the question and then passes the original and rephrased\nquestions together to a different responding LLM. This facilitates the\neffective utilization of rephrased questions generated by one LLM with another.\nOur experiments demonstrate that our methods significantly improve the\nperformance of different models across a wide range to tasks. We further\nprovide a comprehensive comparison between RaR and the popular Chain-of-Thought\n(CoT) methods, both theoretically and empirically. We show that RaR is\ncomplementary to CoT and can be combined with CoT to achieve even better\nperformance. Our work not only contributes to enhancing LLM performance\nefficiently and effectively but also sheds light on a fair evaluation of LLM\ncapabilities. Data and codes are available at\nhttps://github.com/uclaml/Rephrase-and-Respond.",
+        "authors": "Yihe Deng, Weitong Zhang, Zixiang Chen, Quanquan Gu",
+        "published": "2023-11-07",
+        "updated": "2024-04-18",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2401.04057v1",
+        "title": "Unveiling Bias in Fairness Evaluations of Large Language Models: A Critical Literature Review of Music and Movie Recommendation Systems",
+        "abstract": "The rise of generative artificial intelligence, particularly Large Language\nModels (LLMs), has intensified the imperative to scrutinize fairness alongside\naccuracy. Recent studies have begun to investigate fairness evaluations for\nLLMs within domains such as recommendations. Given that personalization is an\nintrinsic aspect of recommendation systems, its incorporation into fairness\nassessments is paramount. Yet, the degree to which current fairness evaluation\nframeworks account for personalization remains unclear. Our comprehensive\nliterature review aims to fill this gap by examining how existing frameworks\nhandle fairness evaluations of LLMs, with a focus on the integration of\npersonalization factors. Despite an exhaustive collection and analysis of\nrelevant works, we discovered that most evaluations overlook personalization, a\ncritical facet of recommendation systems, thereby inadvertently perpetuating\nunfair practices. Our findings shed light on this oversight and underscore the\nurgent need for more nuanced fairness evaluations that acknowledge\npersonalization. Such improvements are vital for fostering equitable\ndevelopment within the AI community.",
+        "authors": "Chandan Kumar Sah, Dr. Lian Xiaoli, Muhammad Mirajul Islam",
+        "published": "2024-01-08",
+        "updated": "2024-01-08",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR",
+            "cs.AI",
+            "cs.SE"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.04814v2",
+        "title": "Evaluation of LLMs on Syntax-Aware Code Fill-in-the-Middle Tasks",
+        "abstract": "We introduce Syntax-Aware Fill-In-the-Middle (SAFIM), a new benchmark for\nevaluating Large Language Models (LLMs) on the code Fill-in-the-Middle (FIM)\ntask. This benchmark focuses on syntax-aware completions of program structures\nsuch as code blocks and conditional expressions, and includes 17,720 examples\nfrom multiple programming languages, sourced from recent code submissions after\nApril 2022 to minimize data contamination. SAFIM provides a robust framework\nwith various prompt designs and novel syntax-aware post-processing techniques,\nfacilitating accurate and fair comparisons across LLMs. Our comprehensive\nevaluation of 15 LLMs shows that FIM pretraining not only enhances FIM\nproficiency but also improves Left-to-Right (L2R) inference using LLMs. Our\nfindings challenge conventional beliefs and suggest that pretraining methods\nand data quality have more impact than model size. SAFIM thus serves as a\nfoundational platform for future research in effective pretraining strategies\nfor code LLMs. The evaluation toolkit and dataset are available at\nhttps://github.com/gonglinyuan/safim, and the leaderboard is available at\nhttps://safimbenchmark.com.",
+        "authors": "Linyuan Gong, Sida Wang, Mostafa Elhoushi, Alvin Cheung",
+        "published": "2024-03-07",
+        "updated": "2024-04-10",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.LG",
+            "cs.SE"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.03514v3",
+        "title": "Can Large Language Models Transform Computational Social Science?",
+        "abstract": "Large Language Models (LLMs) are capable of successfully performing many\nlanguage processing tasks zero-shot (without training data). If zero-shot LLMs\ncan also reliably classify and explain social phenomena like persuasiveness and\npolitical ideology, then LLMs could augment the Computational Social Science\n(CSS) pipeline in important ways. This work provides a road map for using LLMs\nas CSS tools. Towards this end, we contribute a set of prompting best practices\nand an extensive evaluation pipeline to measure the zero-shot performance of 13\nlanguage models on 25 representative English CSS benchmarks. On taxonomic\nlabeling tasks (classification), LLMs fail to outperform the best fine-tuned\nmodels but still achieve fair levels of agreement with humans. On free-form\ncoding tasks (generation), LLMs produce explanations that often exceed the\nquality of crowdworkers' gold references. We conclude that the performance of\ntoday's LLMs can augment the CSS research pipeline in two ways: (1) serving as\nzero-shot data annotators on human annotation teams, and (2) bootstrapping\nchallenging creative generation tasks (e.g., explaining the underlying\nattributes of a text). In summary, LLMs are posed to meaningfully participate\nin social science analysis in partnership with humans.",
+        "authors": "Caleb Ziems, William Held, Omar Shaikh, Jiaao Chen, Zhehao Zhang, Diyi Yang",
+        "published": "2023-04-12",
+        "updated": "2024-02-26",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2308.05345v3",
+        "title": "RTLLM: An Open-Source Benchmark for Design RTL Generation with Large Language Model",
+        "abstract": "Inspired by the recent success of large language models (LLMs) like ChatGPT,\nresearchers start to explore the adoption of LLMs for agile hardware design,\nsuch as generating design RTL based on natural-language instructions. However,\nin existing works, their target designs are all relatively simple and in a\nsmall scale, and proposed by the authors themselves, making a fair comparison\namong different LLM solutions challenging. In addition, many prior works only\nfocus on the design correctness, without evaluating the design qualities of\ngenerated design RTL. In this work, we propose an open-source benchmark named\nRTLLM, for generating design RTL with natural language instructions. To\nsystematically evaluate the auto-generated design RTL, we summarized three\nprogressive goals, named syntax goal, functionality goal, and design quality\ngoal. This benchmark can automatically provide a quantitative evaluation of any\ngiven LLM-based solution. Furthermore, we propose an easy-to-use yet\nsurprisingly effective prompt engineering technique named self-planning, which\nproves to significantly boost the performance of GPT-3.5 in our proposed\nbenchmark.",
+        "authors": "Yao Lu, Shang Liu, Qijun Zhang, Zhiyao Xie",
+        "published": "2023-08-10",
+        "updated": "2023-11-11",
+        "primary_cat": "cs.LG",
+        "cats": [
+            "cs.LG",
+            "cs.AR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2304.03728v1",
+        "title": "Interpretable Unified Language Checking",
+        "abstract": "Despite recent concerns about undesirable behaviors generated by large\nlanguage models (LLMs), including non-factual, biased, and hateful language, we\nfind LLMs are inherent multi-task language checkers based on their latent\nrepresentations of natural and social knowledge. We present an interpretable,\nunified, language checking (UniLC) method for both human and machine-generated\nlanguage that aims to check if language input is factual and fair. While\nfairness and fact-checking tasks have been handled separately with dedicated\nmodels, we find that LLMs can achieve high performance on a combination of\nfact-checking, stereotype detection, and hate speech detection tasks with a\nsimple, few-shot, unified set of prompts. With the ``1/2-shot'' multi-task\nlanguage checking method proposed in this work, the GPT3.5-turbo model\noutperforms fully supervised baselines on several language tasks. The simple\napproach and results suggest that based on strong latent knowledge\nrepresentations, an LLM can be an adaptive and explainable tool for detecting\nmisinformation, stereotypes, and hate speech.",
+        "authors": "Tianhua Zhang, Hongyin Luo, Yung-Sung Chuang, Wei Fang, Luc Gaitskell, Thomas Hartvigsen, Xixin Wu, Danny Fox, Helen Meng, James Glass",
+        "published": "2023-04-07",
+        "updated": "2023-04-07",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.07688v1",
+        "title": "CyberMetric: A Benchmark Dataset for Evaluating Large Language Models Knowledge in Cybersecurity",
+        "abstract": "Large Language Models (LLMs) excel across various domains, from computer\nvision to medical diagnostics. However, understanding the diverse landscape of\ncybersecurity, encompassing cryptography, reverse engineering, and managerial\nfacets like risk assessment, presents a challenge, even for human experts. In\nthis paper, we introduce CyberMetric, a benchmark dataset comprising 10,000\nquestions sourced from standards, certifications, research papers, books, and\nother publications in the cybersecurity domain. The questions are created\nthrough a collaborative process, i.e., merging expert knowledge with LLMs,\nincluding GPT-3.5 and Falcon-180B. Human experts spent over 200 hours verifying\ntheir accuracy and relevance. Beyond assessing LLMs' knowledge, the dataset's\nmain goal is to facilitate a fair comparison between humans and different LLMs\nin cybersecurity. To achieve this, we carefully selected 80 questions covering\na wide range of topics within cybersecurity and involved 30 participants of\ndiverse expertise levels, facilitating a comprehensive comparison between human\nand machine intelligence in this area. The findings revealed that LLMs\noutperformed humans in almost every aspect of cybersecurity.",
+        "authors": "Norbert Tihanyi, Mohamed Amine Ferrag, Ridhi Jain, Merouane Debbah",
+        "published": "2024-02-12",
+        "updated": "2024-02-12",
+        "primary_cat": "cs.AI",
+        "cats": [
+            "cs.AI",
+            "cs.CR"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2401.00588v1",
+        "title": "Fairness in Serving Large Language Models",
+        "abstract": "High-demand LLM inference services (e.g., ChatGPT and BARD) support a wide\nrange of requests from short chat conversations to long document reading. To\nensure that all client requests are processed fairly, most major LLM inference\nservices have request rate limits, to ensure that no client can dominate the\nrequest queue. However, this rudimentary notion of fairness also results in\nunder-utilization of the resources and poor client experience when there is\nspare capacity. While there is a rich literature on fair scheduling, serving\nLLMs presents new challenges due to their unpredictable request lengths and\ntheir unique batching characteristics on parallel accelerators. This paper\nintroduces the definition of LLM serving fairness based on a cost function that\naccounts for the number of input and output tokens processed. To achieve\nfairness in serving, we propose a novel scheduling algorithm, the Virtual Token\nCounter (VTC), a fair scheduler based on the continuous batching mechanism. We\nprove a 2x tight upper bound on the service difference between two backlogged\nclients, adhering to the requirement of work-conserving. Through extensive\nexperiments, we demonstrate the superior performance of VTC in ensuring\nfairness, especially in contrast to other baseline methods, which exhibit\nshortcomings under various conditions.",
+        "authors": "Ying Sheng, Shiyi Cao, Dacheng Li, Banghua Zhu, Zhuohan Li, Danyang Zhuo, Joseph E. Gonzalez, Ion Stoica",
+        "published": "2023-12-31",
+        "updated": "2023-12-31",
+        "primary_cat": "cs.AI",
+        "cats": [
+            "cs.AI",
+            "cs.LG",
+            "cs.PF"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.19118v1",
+        "title": "Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate",
+        "abstract": "Modern large language models (LLMs) like ChatGPT have shown remarkable\nperformance on general language tasks but still struggle on complex reasoning\ntasks, which drives the research on cognitive behaviors of LLMs to explore\nhuman-like problem-solving strategies. Along this direction, one representative\nstrategy is self-reflection, which asks an LLM to refine the solution with the\nfeedback generated by itself iteratively. However, our study shows that such\nreflection-style methods suffer from the Degeneration-of-Thought (DoT) problem:\nonce the LLM has established confidence in its solutions, it is unable to\ngenerate novel thoughts later through reflection even if its initial stance is\nincorrect. To address the DoT problem, we propose a Multi-Agent Debate (MAD)\nframework, in which multiple agents express their arguments in the state of\n\"tit for tat\" and a judge manages the debate process to obtain a final\nsolution. Clearly, our MAD framework encourages divergent thinking in LLMs\nwhich would be helpful for tasks that require deep levels of contemplation.\nExperiment results on two challenging datasets, commonsense machine translation\nand counter-intuitive arithmetic reasoning, demonstrate the effectiveness of\nour MAD framework. Extensive analyses suggest that the adaptive break of debate\nand the modest level of \"tit for tat\" state are required for MAD to obtain good\nperformance. Moreover, we find that LLMs might not be a fair judge if different\nLLMs are used for agents. Codes:\nhttps://github.com/Skytliang/Multi-Agents-Debate",
+        "authors": "Tian Liang, Zhiwei He, Wenxiang Jiao, Xing Wang, Yan Wang, Rui Wang, Yujiu Yang, Zhaopeng Tu, Shuming Shi",
+        "published": "2023-05-30",
+        "updated": "2023-05-30",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.07609v3",
+        "title": "Is ChatGPT Fair for Recommendation? Evaluating Fairness in Large Language Model Recommendation",
+        "abstract": "The remarkable achievements of Large Language Models (LLMs) have led to the\nemergence of a novel recommendation paradigm -- Recommendation via LLM\n(RecLLM). Nevertheless, it is important to note that LLMs may contain social\nprejudices, and therefore, the fairness of recommendations made by RecLLM\nrequires further investigation. To avoid the potential risks of RecLLM, it is\nimperative to evaluate the fairness of RecLLM with respect to various sensitive\nattributes on the user side. Due to the differences between the RecLLM paradigm\nand the traditional recommendation paradigm, it is problematic to directly use\nthe fairness benchmark of traditional recommendation. To address the dilemma,\nwe propose a novel benchmark called Fairness of Recommendation via LLM\n(FaiRLLM). This benchmark comprises carefully crafted metrics and a dataset\nthat accounts for eight sensitive attributes1 in two recommendation scenarios:\nmusic and movies. By utilizing our FaiRLLM benchmark, we conducted an\nevaluation of ChatGPT and discovered that it still exhibits unfairness to some\nsensitive attributes when generating recommendations. Our code and dataset can\nbe found at https://github.com/jizhi-zhang/FaiRLLM.",
+        "authors": "Jizhi Zhang, Keqin Bao, Yang Zhang, Wenjie Wang, Fuli Feng, Xiangnan He",
+        "published": "2023-05-12",
+        "updated": "2023-10-17",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR",
+            "cs.CL",
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2405.01769v1",
+        "title": "A Survey on Large Language Models for Critical Societal Domains: Finance, Healthcare, and Law",
+        "abstract": "In the fast-evolving domain of artificial intelligence, large language models\n(LLMs) such as GPT-3 and GPT-4 are revolutionizing the landscapes of finance,\nhealthcare, and law: domains characterized by their reliance on professional\nexpertise, challenging data acquisition, high-stakes, and stringent regulatory\ncompliance. This survey offers a detailed exploration of the methodologies,\napplications, challenges, and forward-looking opportunities of LLMs within\nthese high-stakes sectors. We highlight the instrumental role of LLMs in\nenhancing diagnostic and treatment methodologies in healthcare, innovating\nfinancial analytics, and refining legal interpretation and compliance\nstrategies. Moreover, we critically examine the ethics for LLM applications in\nthese fields, pointing out the existing ethical concerns and the need for\ntransparent, fair, and robust AI systems that respect regulatory norms. By\npresenting a thorough review of current literature and practical applications,\nwe showcase the transformative impact of LLMs, and outline the imperative for\ninterdisciplinary cooperation, methodological advancements, and ethical\nvigilance. Through this lens, we aim to spark dialogue and inspire future\nresearch dedicated to maximizing the benefits of LLMs while mitigating their\nrisks in these precision-dependent sectors. To facilitate future research on\nLLMs in these critical societal domains, we also initiate a reading list that\ntracks the latest advancements under this topic, which will be continually\nupdated: \\url{https://github.com/czyssrs/LLM_X_papers}.",
+        "authors": "Zhiyu Zoey Chen, Jing Ma, Xinlu Zhang, Nan Hao, An Yan, Armineh Nourbakhsh, Xianjun Yang, Julian McAuley, Linda Petzold, William Yang Wang",
+        "published": "2024-05-02",
+        "updated": "2024-05-02",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2307.15997v1",
+        "title": "RoCar: A Relationship Network-based Evaluation Method to Large Language Models",
+        "abstract": "Large language models (LLMs) have received increasing attention. However, due\nto the complexity of its capabilities, how to rationally evaluate the\ncapabilities of LLMs is still a task to be solved. We propose the RoCar method,\nwhich utilizes the defined basic schemas to randomly construct a task graph and\ngenerates natural language evaluation tasks based on the task graph to evaluate\nthe reasoning and memory abilities of LLMs respectively. Due to the very large\nrandomness of the task construction process, it is possible to ensure that none\nof the LLMs to be tested has directly learned the evaluation tasks,\nguaranteeing the fairness of the evaluation method.",
+        "authors": "Ming Wang, Wenfang Wu, Chongyun Gao, Daling Wang, Shi Feng, Yifei Zhang",
+        "published": "2023-07-29",
+        "updated": "2023-07-29",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.06852v2",
+        "title": "ChemLLM: A Chemical Large Language Model",
+        "abstract": "Large language models (LLMs) have made impressive progress in chemistry\napplications. However, the community lacks an LLM specifically designed for\nchemistry. The main challenges are two-fold: firstly, most chemical data and\nscientific knowledge are stored in structured databases, which limits the\nmodel's ability to sustain coherent dialogue when used directly. Secondly,\nthere is an absence of objective and fair benchmark that encompass most\nchemistry tasks. Here, we introduce ChemLLM, a comprehensive framework that\nfeatures the first LLM dedicated to chemistry. It also includes ChemData, a\ndataset specifically designed for instruction tuning, and ChemBench, a robust\nbenchmark covering nine essential chemistry tasks. ChemLLM is adept at\nperforming various tasks across chemical disciplines with fluid dialogue\ninteraction. Notably, ChemLLM achieves results comparable to GPT-4 on the core\nchemical tasks and demonstrates competitive performance with LLMs of similar\nsize in general scenarios. ChemLLM paves a new path for exploration in chemical\nstudies, and our method of incorporating structured chemical knowledge into\ndialogue systems sets a new standard for developing LLMs in various scientific\nfields. Codes, Datasets, and Model weights are publicly accessible at\nhttps://hf.co/AI4Chem",
+        "authors": "Di Zhang, Wei Liu, Qian Tan, Jingdan Chen, Hang Yan, Yuliang Yan, Jiatong Li, Weiran Huang, Xiangyu Yue, Wanli Ouyang, Dongzhan Zhou, Shufei Zhang, Mao Su, Han-Sen Zhong, Yuqiang Li",
+        "published": "2024-02-10",
+        "updated": "2024-04-25",
+        "primary_cat": "cs.AI",
+        "cats": [
+            "cs.AI",
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.19465v1",
+        "title": "Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Models",
+        "abstract": "Ensuring the trustworthiness of large language models (LLMs) is crucial. Most\nstudies concentrate on fully pre-trained LLMs to better understand and improve\nLLMs' trustworthiness. In this paper, to reveal the untapped potential of\npre-training, we pioneer the exploration of LLMs' trustworthiness during this\nperiod, focusing on five key dimensions: reliability, privacy, toxicity,\nfairness, and robustness. To begin with, we apply linear probing to LLMs. The\nhigh probing accuracy suggests that \\textit{LLMs in early pre-training can\nalready distinguish concepts in each trustworthiness dimension}. Therefore, to\nfurther uncover the hidden possibilities of pre-training, we extract steering\nvectors from a LLM's pre-training checkpoints to enhance the LLM's\ntrustworthiness. Finally, inspired by~\\citet{choi2023understanding} that mutual\ninformation estimation is bounded by linear probing accuracy, we also probe\nLLMs with mutual information to investigate the dynamics of trustworthiness\nduring pre-training. We are the first to observe a similar two-phase\nphenomenon: fitting and compression~\\citep{shwartz2017opening}. This research\nprovides an initial exploration of trustworthiness modeling during LLM\npre-training, seeking to unveil new insights and spur further developments in\nthe field. We will make our code publicly accessible at\n\\url{https://github.com/ChnQ/TracingLLM}.",
+        "authors": "Chen Qian, Jie Zhang, Wei Yao, Dongrui Liu, Zhenfei Yin, Yu Qiao, Yong Liu, Jing Shao",
+        "published": "2024-02-29",
+        "updated": "2024-02-29",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.08189v1",
+        "title": "Simulating Human Strategic Behavior: Comparing Single and Multi-agent LLMs",
+        "abstract": "When creating plans, policies, or applications for people, it is challenging\nfor designers to think through the strategic ways that different people will\nbehave. Recently, Large Language Models (LLMs) have been shown to create\nrealistic simulations of human-like behavior based on personas. We build on\nthis to investigate whether LLMs can simulate human strategic behavior. Human\nstrategies are complex because they take into account social norms in addition\nto aiming to maximize personal gain. The ultimatum game is a classic economics\nexperiment used to understand human strategic behavior in a social setting. It\nshows that people will often choose to \"punish\" other players to enforce social\nnorms rather than to maximize personal profits. We test whether LLMs can\nreplicate this complex behavior in simulations. We compare two architectures:\nsingle- and multi-agent LLMs. We compare their abilities to (1) simulate\nhuman-like actions in the ultimatum game, (2) simulate two player\npersonalities, greedy and fair, and (3) create robust strategies that are\nlogically complete and consistent with personality. Our evaluation shows the\nmulti-agent architecture is much more accurate than single LLMs (88% vs. 50%)\nin simulating human strategy creation and actions for personality pairs. Thus\nthere is potential to use LLMs to simulate human strategic behavior to help\ndesigners, planners, and policymakers perform preliminary exploration of how\npeople behave in systems.",
+        "authors": "Karthik Sreedhar, Lydia Chilton",
+        "published": "2024-02-13",
+        "updated": "2024-02-13",
+        "primary_cat": "cs.HC",
+        "cats": [
+            "cs.HC"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.17553v1",
+        "title": "RuBia: A Russian Language Bias Detection Dataset",
+        "abstract": "Warning: this work contains upsetting or disturbing content.\n  Large language models (LLMs) tend to learn the social and cultural biases\npresent in the raw pre-training data. To test if an LLM's behavior is fair,\nfunctional datasets are employed, and due to their purpose, these datasets are\nhighly language and culture-specific. In this paper, we address a gap in the\nscope of multilingual bias evaluation by presenting a bias detection dataset\nspecifically designed for the Russian language, dubbed as RuBia. The RuBia\ndataset is divided into 4 domains: gender, nationality, socio-economic status,\nand diverse, each of the domains is further divided into multiple fine-grained\nsubdomains. Every example in the dataset consists of two sentences with the\nfirst reinforcing a potentially harmful stereotype or trope and the second\ncontradicting it. These sentence pairs were first written by volunteers and\nthen validated by native-speaking crowdsourcing workers. Overall, there are\nnearly 2,000 unique sentence pairs spread over 19 subdomains in RuBia. To\nillustrate the dataset's purpose, we conduct a diagnostic evaluation of\nstate-of-the-art or near-state-of-the-art LLMs and discuss the LLMs'\npredisposition to social biases.",
+        "authors": "Veronika Grigoreva, Anastasiia Ivanova, Ilseyar Alimova, Ekaterina Artemova",
+        "published": "2024-03-26",
+        "updated": "2024-03-26",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.04489v1",
+        "title": "De-amplifying Bias from Differential Privacy in Language Model Fine-tuning",
+        "abstract": "Fairness and privacy are two important values machine learning (ML)\npractitioners often seek to operationalize in models. Fairness aims to reduce\nmodel bias for social/demographic sub-groups. Privacy via differential privacy\n(DP) mechanisms, on the other hand, limits the impact of any individual's\ntraining data on the resulting model. The trade-offs between privacy and\nfairness goals of trustworthy ML pose a challenge to those wishing to address\nboth. We show that DP amplifies gender, racial, and religious bias when\nfine-tuning large language models (LLMs), producing models more biased than\nones fine-tuned without DP. We find the cause of the amplification to be a\ndisparity in convergence of gradients across sub-groups. Through the case of\nbinary gender bias, we demonstrate that Counterfactual Data Augmentation (CDA),\na known method for addressing bias, also mitigates bias amplification by DP. As\na consequence, DP and CDA together can be used to fine-tune models while\nmaintaining both fairness and privacy.",
+        "authors": "Sanjari Srivastava, Piotr Mardziel, Zhikhun Zhang, Archana Ahlawat, Anupam Datta, John C Mitchell",
+        "published": "2024-02-07",
+        "updated": "2024-02-07",
+        "primary_cat": "cs.LG",
+        "cats": [
+            "cs.LG",
+            "cs.CR",
+            "cs.CY",
+            "stat.ME"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2312.15478v1",
+        "title": "A Group Fairness Lens for Large Language Models",
+        "abstract": "The rapid advancement of large language models has revolutionized various\napplications but also raised crucial concerns about their potential to\nperpetuate biases and unfairness when deployed in social media contexts.\nEvaluating LLMs' potential biases and fairness has become crucial, as existing\nmethods rely on limited prompts focusing on just a few groups, lacking a\ncomprehensive categorical perspective. In this paper, we propose evaluating LLM\nbiases from a group fairness lens using a novel hierarchical schema\ncharacterizing diverse social groups. Specifically, we construct a dataset,\nGFair, encapsulating target-attribute combinations across multiple dimensions.\nIn addition, we introduce statement organization, a new open-ended text\ngeneration task, to uncover complex biases in LLMs. Extensive evaluations of\npopular LLMs reveal inherent safety concerns. To mitigate the biases of LLM\nfrom a group fairness perspective, we pioneer a novel chain-of-thought method\nGF-Think to mitigate biases of LLMs from a group fairness perspective.\nExperimental results demonstrate its efficacy in mitigating bias in LLMs to\nachieve fairness.",
+        "authors": "Guanqun Bi, Lei Shen, Yuqiang Xie, Yanan Cao, Tiangang Zhu, Xiaodong He",
+        "published": "2023-12-24",
+        "updated": "2023-12-24",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2312.15198v2",
+        "title": "Do LLM Agents Exhibit Social Behavior?",
+        "abstract": "The advances of Large Language Models (LLMs) are expanding their utility in\nboth academic research and practical applications. Recent social science\nresearch has explored the use of these ``black-box'' LLM agents for simulating\ncomplex social systems and potentially substituting human subjects in\nexperiments. Our study delves into this emerging domain, investigating the\nextent to which LLMs exhibit key social interaction principles, such as social\nlearning, social preference, and cooperative behavior (indirect reciprocity),\nin their interactions with humans and other agents. We develop a framework for\nour study, wherein classical laboratory experiments involving human subjects\nare adapted to use LLM agents. This approach involves step-by-step reasoning\nthat mirrors human cognitive processes and zero-shot learning to assess the\ninnate preferences of LLMs. Our analysis of LLM agents' behavior includes both\nthe primary effects and an in-depth examination of the underlying mechanisms.\nFocusing on GPT-4, our analyses suggest that LLM agents appear to exhibit a\nrange of human-like social behaviors such as distributional and reciprocity\npreferences, responsiveness to group identity cues, engagement in indirect\nreciprocity, and social learning capabilities. However, our analysis also\nreveals notable differences: LLMs demonstrate a pronounced fairness preference,\nweaker positive reciprocity, and a more calculating approach in social learning\ncompared to humans. These insights indicate that while LLMs hold great promise\nfor applications in social science research, such as in laboratory experiments\nand agent-based modeling, the subtle behavioral differences between LLM agents\nand humans warrant further investigation. Careful examination and development\nof protocols in evaluating the social behaviors of LLMs are necessary before\ndirectly applying these models to emulate human behavior.",
+        "authors": "Yan Leng, Yuan Yuan",
+        "published": "2023-12-23",
+        "updated": "2024-02-22",
+        "primary_cat": "cs.AI",
+        "cats": [
+            "cs.AI",
+            "cs.SI",
+            "econ.GN",
+            "q-fin.EC"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2312.06056v1",
+        "title": "METAL: Metamorphic Testing Framework for Analyzing Large-Language Model Qualities",
+        "abstract": "Large-Language Models (LLMs) have shifted the paradigm of natural language\ndata processing. However, their black-boxed and probabilistic characteristics\ncan lead to potential risks in the quality of outputs in diverse LLM\napplications. Recent studies have tested Quality Attributes (QAs), such as\nrobustness or fairness, of LLMs by generating adversarial input texts. However,\nexisting studies have limited their coverage of QAs and tasks in LLMs and are\ndifficult to extend. Additionally, these studies have only used one evaluation\nmetric, Attack Success Rate (ASR), to assess the effectiveness of their\napproaches. We propose a MEtamorphic Testing for Analyzing LLMs (METAL)\nframework to address these issues by applying Metamorphic Testing (MT)\ntechniques. This approach facilitates the systematic testing of LLM qualities\nby defining Metamorphic Relations (MRs), which serve as modularized evaluation\nmetrics. The METAL framework can automatically generate hundreds of MRs from\ntemplates that cover various QAs and tasks. In addition, we introduced novel\nmetrics that integrate the ASR method into the semantic qualities of text to\nassess the effectiveness of MRs accurately. Through the experiments conducted\nwith three prominent LLMs, we have confirmed that the METAL framework\neffectively evaluates essential QAs on primary LLM tasks and reveals the\nquality risks in LLMs. Moreover, the newly proposed metrics can guide the\noptimal MRs for testing each task and suggest the most effective method for\ngenerating MRs.",
+        "authors": "Sangwon Hyun, Mingyu Guo, M. Ali Babar",
+        "published": "2023-12-11",
+        "updated": "2023-12-11",
+        "primary_cat": "cs.SE",
+        "cats": [
+            "cs.SE",
+            "cs.AI",
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2308.11483v1",
+        "title": "Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questions",
+        "abstract": "Large Language Models (LLMs) have demonstrated remarkable capabilities in\nvarious NLP tasks. However, previous works have shown these models are\nsensitive towards prompt wording, and few-shot demonstrations and their order,\nposing challenges to fair assessment of these models. As these models become\nmore powerful, it becomes imperative to understand and address these\nlimitations. In this paper, we focus on LLMs robustness on the task of\nmultiple-choice questions -- commonly adopted task to study reasoning and\nfact-retrieving capability of LLMs. Investigating the sensitivity of LLMs\ntowards the order of options in multiple-choice questions, we demonstrate a\nconsiderable performance gap of approximately 13% to 75% in LLMs on different\nbenchmarks, when answer options are reordered, even when using demonstrations\nin a few-shot setting. Through a detailed analysis, we conjecture that this\nsensitivity arises when LLMs are uncertain about the prediction between the\ntop-2/3 choices, and specific options placements may favor certain prediction\nbetween those top choices depending on the question caused by positional bias.\nWe also identify patterns in top-2 choices that amplify or mitigate the model's\nbias toward option placement. We found that for amplifying bias, the optimal\nstrategy involves positioning the top two choices as the first and last\noptions. Conversely, to mitigate bias, we recommend placing these choices among\nthe adjacent options. To validate our conjecture, we conduct various\nexperiments and adopt two approaches to calibrate LLMs' predictions, leading to\nup to 8 percentage points improvement across different models and benchmarks.",
+        "authors": "Pouya Pezeshkpour, Estevam Hruschka",
+        "published": "2023-08-22",
+        "updated": "2023-08-22",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2309.14345v2",
+        "title": "Bias Testing and Mitigation in LLM-based Code Generation",
+        "abstract": "Utilizing state-of-the-art Large Language Models (LLMs), automatic code\ngeneration models play a pivotal role in enhancing the productivity of software\ndevelopment procedures. As the adoption of LLMs becomes more widespread in\nsoftware coding ecosystems, a pressing issue has emerged: does the generated\ncode contain social bias and unfairness, such as those related to age, gender,\nand race? This issue concerns the integrity, fairness, and ethical foundation\nof software applications that depend on the code generated by these models, yet\nis under-explored in the literature. This paper presents a novel bias testing\nframework that is specifically designed for code generation tasks. Based on\nthis framework, we conduct an extensive evaluation of the bias in code\ngenerated by five state-of-the-art LLMs. Our findings reveal that 20.29% to\n44.93% code functions generated by the models under study are biased when\nhandling bias sensitive tasks (i.e., tasks that involve sensitive attributes\nsuch as age and gender). This indicates that the existing LLMs can be unfair in\ncode generation, posing risks of unintended and harmful software behaviors. To\nmitigate bias for code generation models, we evaluate five bias mitigation\nprompt strategies, i.e., utilizing bias testing results to refine the code\n(zero-shot), one-, few-shot, and two Chain-of-Thought (CoT) prompts. Our\nevaluation results illustrate that these strategies are all effective in\nmitigating bias. Overall, one-shot and few-shot learning are the two most\neffective. For GPT-4, 80% to 90% code bias can be removed with one-shot\nlearning.",
+        "authors": "Dong Huang, Qingwen Bu, Jie Zhang, Xiaofei Xie, Junjie Chen, Heming Cui",
+        "published": "2023-09-03",
+        "updated": "2024-01-09",
+        "primary_cat": "cs.SE",
+        "cats": [
+            "cs.SE",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2303.01248v3",
+        "title": "Can ChatGPT Assess Human Personalities? A General Evaluation Framework",
+        "abstract": "Large Language Models (LLMs) especially ChatGPT have produced impressive\nresults in various areas, but their potential human-like psychology is still\nlargely unexplored. Existing works study the virtual personalities of LLMs but\nrarely explore the possibility of analyzing human personalities via LLMs. This\npaper presents a generic evaluation framework for LLMs to assess human\npersonalities based on Myers Briggs Type Indicator (MBTI) tests. Specifically,\nwe first devise unbiased prompts by randomly permuting options in MBTI\nquestions and adopt the average testing result to encourage more impartial\nanswer generation. Then, we propose to replace the subject in question\nstatements to enable flexible queries and assessments on different subjects\nfrom LLMs. Finally, we re-formulate the question instructions in a manner of\ncorrectness evaluation to facilitate LLMs to generate clearer responses. The\nproposed framework enables LLMs to flexibly assess personalities of different\ngroups of people. We further propose three evaluation metrics to measure the\nconsistency, robustness, and fairness of assessment results from\nstate-of-the-art LLMs including ChatGPT and GPT-4. Our experiments reveal\nChatGPT's ability to assess human personalities, and the average results\ndemonstrate that it can achieve more consistent and fairer assessments in spite\nof lower robustness against prompt biases compared with InstructGPT.",
+        "authors": "Haocong Rao, Cyril Leung, Chunyan Miao",
+        "published": "2023-03-01",
+        "updated": "2023-10-13",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.17916v2",
+        "title": "LLM-Resistant Math Word Problem Generation via Adversarial Attacks",
+        "abstract": "Large language models (LLMs) have significantly transformed the educational\nlandscape. As current plagiarism detection tools struggle to keep pace with\nLLMs' rapid advancements, the educational community faces the challenge of\nassessing students' true problem-solving abilities in the presence of LLMs. In\nthis work, we explore a new paradigm for ensuring fair evaluation -- generating\nadversarial examples which preserve the structure and difficulty of the\noriginal questions aimed for assessment, but are unsolvable by LLMs. Focusing\non the domain of math word problems, we leverage abstract syntax trees to\nstructurally generate adversarial examples that cause LLMs to produce incorrect\nanswers by simply editing the numeric values in the problems. We conduct\nexperiments on various open- and closed-source LLMs, quantitatively and\nqualitatively demonstrating that our method significantly degrades their math\nproblem-solving ability. We identify shared vulnerabilities among LLMs and\npropose a cost-effective approach to attack high-cost models. Additionally, we\nconduct automatic analysis on math problems and investigate the cause of\nfailure, offering a nuanced view into model's limitation.",
+        "authors": "Roy Xie, Chengxuan Huang, Junlin Wang, Bhuwan Dhingra",
+        "published": "2024-02-27",
+        "updated": "2024-03-30",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.12090v1",
+        "title": "UP5: Unbiased Foundation Model for Fairness-aware Recommendation",
+        "abstract": "Recent advancements in foundation models such as large language models (LLM)\nhave propelled them to the forefront of recommender systems (RS). Moreover,\nfairness in RS is critical since many users apply it for decision-making and\ndemand fulfillment. However, at present, there is a lack of understanding\nregarding the level of fairness exhibited by recommendation foundation models\nand the appropriate methods for equitably treating different groups of users in\nfoundation models. In this paper, we focus on user-side unfairness problem and\nshow through a thorough examination that there is unfairness involved in LLMs\nthat lead to unfair recommendation results. To eliminate bias from LLM for\nfairness-aware recommendation, we introduce a novel Unbiased P5 (UP5)\nfoundation model based on Counterfactually-Fair-Prompting (CFP) techniques. CFP\nincludes two sub-modules: a personalized prefix prompt that enhances fairness\nwith respect to individual sensitive attributes, and a Prompt Mixture that\nintegrates multiple counterfactually-fair prompts for a set of sensitive\nattributes. Experiments are conducted on two real-world datasets, MovieLens-1M\nand Insurance, and results are compared with both matching-based and\nsequential-based fairness-aware recommendation models. The results show that\nUP5 achieves better recommendation performance and meanwhile exhibits a high\nlevel of fairness.",
+        "authors": "Wenyue Hua, Yingqiang Ge, Shuyuan Xu, Jianchao Ji, Yongfeng Zhang",
+        "published": "2023-05-20",
+        "updated": "2023-05-20",
+        "primary_cat": "cs.IR",
+        "cats": [
+            "cs.IR",
+            "cs.AI",
+            "cs.CL",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.15451v1",
+        "title": "Towards Enabling FAIR Dataspaces Using Large Language Models",
+        "abstract": "Dataspaces have recently gained adoption across various sectors, including\ntraditionally less digitized domains such as culture. Leveraging Semantic Web\ntechnologies helps to make dataspaces FAIR, but their complexity poses a\nsignificant challenge to the adoption of dataspaces and increases their cost.\nThe advent of Large Language Models (LLMs) raises the question of how these\nmodels can support the adoption of FAIR dataspaces. In this work, we\ndemonstrate the potential of LLMs in dataspaces with a concrete example. We\nalso derive a research agenda for exploring this emerging field.",
+        "authors": "Benedikt T. Arnold, Johannes Theissen-Lipp, Diego Collarana, Christoph Lange, Sandra Geisler, Edward Curry, Stefan Decker",
+        "published": "2024-03-18",
+        "updated": "2024-03-18",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2309.09397v1",
+        "title": "Do Large GPT Models Discover Moral Dimensions in Language Representations? A Topological Study Of Sentence Embeddings",
+        "abstract": "As Large Language Models are deployed within Artificial Intelligence systems,\nthat are increasingly integrated with human society, it becomes more important\nthan ever to study their internal structures. Higher level abilities of LLMs\nsuch as GPT-3.5 emerge in large part due to informative language\nrepresentations they induce from raw text data during pre-training on trillions\nof words. These embeddings exist in vector spaces of several thousand\ndimensions, and their processing involves mapping between multiple vector\nspaces, with total number of parameters on the order of trillions. Furthermore,\nthese language representations are induced by gradient optimization, resulting\nin a black box system that is hard to interpret. In this paper, we take a look\nat the topological structure of neuronal activity in the \"brain\" of Chat-GPT's\nfoundation language model, and analyze it with respect to a metric representing\nthe notion of fairness. We develop a novel approach to visualize GPT's moral\ndimensions. We first compute a fairness metric, inspired by social psychology\nliterature, to identify factors that typically influence fairness assessments\nin humans, such as legitimacy, need, and responsibility. Subsequently, we\nsummarize the manifold's shape using a lower-dimensional simplicial complex,\nwhose topology is derived from this metric. We color it with a heat map\nassociated with this fairness metric, producing human-readable visualizations\nof the high-dimensional sentence manifold. Our results show that sentence\nembeddings based on GPT-3.5 can be decomposed into two submanifolds\ncorresponding to fair and unfair moral judgments. This indicates that GPT-based\nlanguage models develop a moral dimension within their representation spaces\nand induce an understanding of fairness during their training process.",
+        "authors": "Stephen Fitz",
+        "published": "2023-09-17",
+        "updated": "2023-09-17",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "cs.CY",
+            "cs.LG",
+            "cs.NE"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2308.10397v2",
+        "title": "FairMonitor: A Four-Stage Automatic Framework for Detecting Stereotypes and Biases in Large Language Models",
+        "abstract": "Detecting stereotypes and biases in Large Language Models (LLMs) can enhance\nfairness and reduce adverse impacts on individuals or groups when these LLMs\nare applied. However, the majority of existing methods focus on measuring the\nmodel's preference towards sentences containing biases and stereotypes within\ndatasets, which lacks interpretability and cannot detect implicit biases and\nstereotypes in the real world. To address this gap, this paper introduces a\nfour-stage framework to directly evaluate stereotypes and biases in the\ngenerated content of LLMs, including direct inquiry testing, serial or adapted\nstory testing, implicit association testing, and unknown situation testing.\nAdditionally, the paper proposes multi-dimensional evaluation metrics and\nexplainable zero-shot prompts for automated evaluation. Using the education\nsector as a case study, we constructed the Edu-FairMonitor based on the\nfour-stage framework, which encompasses 12,632 open-ended questions covering\nnine sensitive factors and 26 educational scenarios. Experimental results\nreveal varying degrees of stereotypes and biases in five LLMs evaluated on\nEdu-FairMonitor. Moreover, the results of our proposed automated evaluation\nmethod have shown a high correlation with human annotations.",
+        "authors": "Yanhong Bai, Jiabao Zhao, Jinxin Shi, Tingjiang Wei, Xingjiao Wu, Liang He",
+        "published": "2023-08-21",
+        "updated": "2023-10-27",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.01937v1",
+        "title": "Can Large Language Models Be an Alternative to Human Evaluations?",
+        "abstract": "Human evaluation is indispensable and inevitable for assessing the quality of\ntexts generated by machine learning models or written by humans. However, human\nevaluation is very difficult to reproduce and its quality is notoriously\nunstable, hindering fair comparisons among different natural language\nprocessing (NLP) models and algorithms. Recently, large language models (LLMs)\nhave demonstrated exceptional performance on unseen tasks when only the task\ninstructions are provided. In this paper, we explore if such an ability of the\nLLMs can be used as an alternative to human evaluation. We present the LLMs\nwith the exact same instructions, samples to be evaluated, and questions used\nto conduct human evaluation, and then ask the LLMs to generate responses to\nthose questions; we dub this LLM evaluation. We use human evaluation and LLM\nevaluation to evaluate the texts in two NLP tasks: open-ended story generation\nand adversarial attacks. We show that the result of LLM evaluation is\nconsistent with the results obtained by expert human evaluation: the texts\nrated higher by human experts are also rated higher by the LLMs. We also find\nthat the results of LLM evaluation are stable over different formatting of the\ntask instructions and the sampling algorithm used to generate the answer. We\nare the first to show the potential of using LLMs to assess the quality of\ntexts and discuss the limitations and ethical considerations of LLM evaluation.",
+        "authors": "Cheng-Han Chiang, Hung-yi Lee",
+        "published": "2023-05-03",
+        "updated": "2023-05-03",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.HC"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2308.10149v2",
+        "title": "A Survey on Fairness in Large Language Models",
+        "abstract": "Large Language Models (LLMs) have shown powerful performance and development\nprospects and are widely deployed in the real world. However, LLMs can capture\nsocial biases from unprocessed training data and propagate the biases to\ndownstream tasks. Unfair LLM systems have undesirable social impacts and\npotential harms. In this paper, we provide a comprehensive review of related\nresearch on fairness in LLMs. Considering the influence of parameter magnitude\nand training paradigm on research strategy, we divide existing fairness\nresearch into oriented to medium-sized LLMs under pre-training and fine-tuning\nparadigms and oriented to large-sized LLMs under prompting paradigms. First,\nfor medium-sized LLMs, we introduce evaluation metrics and debiasing methods\nfrom the perspectives of intrinsic bias and extrinsic bias, respectively. Then,\nfor large-sized LLMs, we introduce recent fairness research, including fairness\nevaluation, reasons for bias, and debiasing methods. Finally, we discuss and\nprovide insight on the challenges and future directions for the development of\nfairness in LLMs.",
+        "authors": "Yingji Li, Mengnan Du, Rui Song, Xin Wang, Ying Wang",
+        "published": "2023-08-20",
+        "updated": "2024-02-21",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2309.03852v2",
+        "title": "FLM-101B: An Open LLM and How to Train It with $100K Budget",
+        "abstract": "Large language models (LLMs) have achieved remarkable success in NLP and\nmultimodal tasks, among others. Despite these successes, two main challenges\nremain in developing LLMs: (i) high computational cost, and (ii) fair and\nobjective evaluations. In this paper, we report a solution to significantly\nreduce LLM training cost through a growth strategy. We demonstrate that a\n101B-parameter LLM with 0.31T tokens can be trained with a budget of 100K US\ndollars. Inspired by IQ tests, we also consolidate an additional range of\nevaluations on top of existing evaluations that focus on knowledge-oriented\nabilities. These IQ evaluations include symbolic mapping, rule understanding,\npattern mining, and anti-interference. Such evaluations minimize the potential\nimpact of memorization. Experimental results show that our model, named\nFLM-101B, trained with a budget of 100K US dollars, achieves performance\ncomparable to powerful and well-known models, e.g., GPT-3 and GLM-130B,\nespecially on the additional range of IQ evaluations. The checkpoint of\nFLM-101B is released at https://huggingface.co/CofeAI/FLM-101B.",
+        "authors": "Xiang Li, Yiqun Yao, Xin Jiang, Xuezhi Fang, Xuying Meng, Siqi Fan, Peng Han, Jing Li, Li Du, Bowen Qin, Zheng Zhang, Aixin Sun, Yequan Wang",
+        "published": "2023-09-07",
+        "updated": "2023-09-17",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2310.18130v2",
+        "title": "DELPHI: Data for Evaluating LLMs' Performance in Handling Controversial Issues",
+        "abstract": "Controversy is a reflection of our zeitgeist, and an important aspect to any\ndiscourse. The rise of large language models (LLMs) as conversational systems\nhas increased public reliance on these systems for answers to their various\nquestions. Consequently, it is crucial to systematically examine how these\nmodels respond to questions that pertaining to ongoing debates. However, few\nsuch datasets exist in providing human-annotated labels reflecting the\ncontemporary discussions. To foster research in this area, we propose a novel\nconstruction of a controversial questions dataset, expanding upon the publicly\nreleased Quora Question Pairs Dataset. This dataset presents challenges\nconcerning knowledge recency, safety, fairness, and bias. We evaluate different\nLLMs using a subset of this dataset, illuminating how they handle controversial\nissues and the stances they adopt. This research ultimately contributes to our\nunderstanding of LLMs' interaction with controversial issues, paving the way\nfor improvements in their comprehension and handling of complex societal\ndebates.",
+        "authors": "David Q. Sun, Artem Abzaliev, Hadas Kotek, Zidi Xiu, Christopher Klein, Jason D. Williams",
+        "published": "2023-10-27",
+        "updated": "2023-11-07",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.HC"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2403.15491v1",
+        "title": "Open Source Conversational LLMs do not know most Spanish words",
+        "abstract": "The growing interest in Large Language Models (LLMs) and in particular in\nconversational models with which users can interact has led to the development\nof a large number of open-source chat LLMs. These models are evaluated on a\nwide range of benchmarks to assess their capabilities in answering questions or\nsolving problems on almost any possible topic or to test their ability to\nreason or interpret texts. Instead, the evaluation of the knowledge that these\nmodels have of the languages has received much less attention. For example, the\nwords that they can recognize and use in different languages. In this paper, we\nevaluate the knowledge that open-source chat LLMs have of Spanish words by\ntesting a sample of words in a reference dictionary. The results show that\nopen-source chat LLMs produce incorrect meanings for an important fraction of\nthe words and are not able to use most of the words correctly to write\nsentences with context. These results show how Spanish is left behind in the\nopen-source LLM race and highlight the need to push for linguistic fairness in\nconversational LLMs ensuring that they provide similar performance across\nlanguages.",
+        "authors": "Javier Conde, Miguel Gonz\u00e1lez, Nina Melero, Raquel Ferrando, Gonzalo Mart\u00ednez, Elena Merino-G\u00f3mez, Jos\u00e9 Alberto Hern\u00e1ndez, Pedro Reviriego",
+        "published": "2024-03-21",
+        "updated": "2024-03-21",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.13925v1",
+        "title": "MARIO Eval: Evaluate Your Math LLM with your Math LLM--A mathematical dataset evaluation toolkit",
+        "abstract": "Large language models (LLMs) have been explored in a variety of reasoning\ntasks including solving of mathematical problems. Each math dataset typically\nincludes its own specially designed evaluation script, which, while suitable\nfor its intended use, lacks generalizability across different datasets.\nConsequently, updates and adaptations to these evaluation tools tend to occur\nwithout being systematically reported, leading to inconsistencies and obstacles\nto fair comparison across studies. To bridge this gap, we introduce a\ncomprehensive mathematical evaluation toolkit that not only utilizes a python\ncomputer algebra system (CAS) for its numerical accuracy, but also integrates\nan optional LLM, known for its considerable natural language processing\ncapabilities. To validate the effectiveness of our toolkit, we manually\nannotated two distinct datasets. Our experiments demonstrate that the toolkit\nyields more robust evaluation results compared to prior works, even without an\nLLM. Furthermore, when an LLM is incorporated, there is a notable enhancement.\nThe code for our method will be made available at\n\\url{https://github.com/MARIO-Math-Reasoning/math_evaluation}.",
+        "authors": "Boning Zhang, Chengxi Li, Kai Fan",
+        "published": "2024-04-22",
+        "updated": "2024-04-22",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.08656v1",
+        "title": "Linear Cross-document Event Coreference Resolution with X-AMR",
+        "abstract": "Event Coreference Resolution (ECR) as a pairwise mention classification task\nis expensive both for automated systems and manual annotations. The task's\nquadratic difficulty is exacerbated when using Large Language Models (LLMs),\nmaking prompt engineering for ECR prohibitively costly. In this work, we\npropose a graphical representation of events, X-AMR, anchored around individual\nmentions using a \\textbf{cross}-document version of \\textbf{A}bstract\n\\textbf{M}eaning \\textbf{R}epresentation. We then linearize the ECR with a\nnovel multi-hop coreference algorithm over the event graphs. The event graphs\nsimplify ECR, making it a) LLM cost-effective, b) compositional and\ninterpretable, and c) easily annotated. For a fair assessment, we first enrich\nan existing ECR benchmark dataset with these event graphs using an\nannotator-friendly tool we introduce. Then, we employ GPT-4, the newest LLM by\nOpenAI, for these annotations. Finally, using the ECR algorithm, we assess\nGPT-4 against humans and analyze its limitations. Through this research, we aim\nto advance the state-of-the-art for efficient ECR and shed light on the\npotential shortcomings of current LLMs at this task. Code and annotations:\n\\url{https://github.com/ahmeshaf/gpt_coref}",
+        "authors": "Shafiuddin Rehan Ahmed, George Arthur Baker, Evi Judge, Michael Regan, Kristin Wright-Bettner, Martha Palmer, James H. Martin",
+        "published": "2024-03-25",
+        "updated": "2024-03-25",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2305.13862v2",
+        "title": "A Trip Towards Fairness: Bias and De-Biasing in Large Language Models",
+        "abstract": "Cheap-to-Build Very Large-Language Models (CtB-LLMs) with affordable training\nare emerging as the next big revolution in natural language processing and\nunderstanding. These CtB-LLMs are democratizing access to trainable Very\nLarge-Language Models (VLLMs) and, thus, may represent the building blocks of\nmany NLP systems solving downstream tasks. Hence, a little or a large bias in\nCtB-LLMs may cause huge harm. In this paper, we performed a large investigation\nof the bias of three families of CtB-LLMs, and we showed that debiasing\ntechniques are effective and usable. Indeed, according to current tests, the\nLLaMA and the OPT families have an important bias in gender, race, religion,\nand profession. In contrast to the analysis for other LLMs, we discovered that\nbias depends not on the number of parameters but on the perplexity. Finally,\nthe debiasing of OPT using LoRA reduces bias up to 4.12 points in the\nnormalized stereotype score.",
+        "authors": "Leonardo Ranaldi, Elena Sofia Ruzzetti, Davide Venditti, Dario Onorati, Fabio Massimo Zanzotto",
+        "published": "2023-05-23",
+        "updated": "2023-08-29",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2312.07420v1",
+        "title": "FairSISA: Ensemble Post-Processing to Improve Fairness of Unlearning in LLMs",
+        "abstract": "Training large language models (LLMs) is a costly endeavour in terms of time\nand computational resources. The large amount of training data used during the\nunsupervised pre-training phase makes it difficult to verify all data and,\nunfortunately, undesirable data may be ingested during training. Re-training\nfrom scratch is impractical and has led to the creation of the 'unlearning'\ndiscipline where models are modified to \"unlearn\" undesirable information\nwithout retraining. However, any modification can alter the behaviour of LLMs,\nespecially on key dimensions such as fairness. This is the first work that\nexamines this interplay between unlearning and fairness for LLMs. In\nparticular, we focus on a popular unlearning framework known as SISA [Bourtoule\net al., 2021], which creates an ensemble of models trained on disjoint shards.\nWe evaluate the performance-fairness trade-off for SISA, and empirically\ndemsontrate that SISA can indeed reduce fairness in LLMs. To remedy this, we\npropose post-processing bias mitigation techniques for ensemble models produced\nby SISA. We adapt the post-processing fairness improvement technique from\n[Hardt et al., 2016] to design three methods that can handle model ensembles,\nand prove that one of the methods is an optimal fair predictor for ensemble of\nmodels. Through experimental results, we demonstrate the efficacy of our\npost-processing framework called 'FairSISA'.",
+        "authors": "Swanand Ravindra Kadhe, Anisa Halimi, Ambrish Rawat, Nathalie Baracaldo",
+        "published": "2023-12-12",
+        "updated": "2023-12-12",
+        "primary_cat": "cs.LG",
+        "cats": [
+            "cs.LG",
+            "cs.CY"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.08517v1",
+        "title": "Online Safety Analysis for LLMs: a Benchmark, an Assessment, and a Path Forward",
+        "abstract": "While Large Language Models (LLMs) have seen widespread applications across\nnumerous fields, their limited interpretability poses concerns regarding their\nsafe operations from multiple aspects, e.g., truthfulness, robustness, and\nfairness. Recent research has started developing quality assurance methods for\nLLMs, introducing techniques such as offline detector-based or uncertainty\nestimation methods. However, these approaches predominantly concentrate on\npost-generation analysis, leaving the online safety analysis for LLMs during\nthe generation phase an unexplored area. To bridge this gap, we conduct in this\nwork a comprehensive evaluation of the effectiveness of existing online safety\nanalysis methods on LLMs. We begin with a pilot study that validates the\nfeasibility of detecting unsafe outputs in the early generation process.\nFollowing this, we establish the first publicly available benchmark of online\nsafety analysis for LLMs, including a broad spectrum of methods, models, tasks,\ndatasets, and evaluation metrics. Utilizing this benchmark, we extensively\nanalyze the performance of state-of-the-art online safety analysis methods on\nboth open-source and closed-source LLMs. This analysis reveals the strengths\nand weaknesses of individual methods and offers valuable insights into\nselecting the most appropriate method based on specific application scenarios\nand task requirements. Furthermore, we also explore the potential of using\nhybridization methods, i.e., combining multiple methods to derive a collective\nsafety conclusion, to enhance the efficacy of online safety analysis for LLMs.\nOur findings indicate a promising direction for the development of innovative\nand trustworthy quality assurance methodologies for LLMs, facilitating their\nreliable deployments across diverse domains.",
+        "authors": "Xuan Xie, Jiayang Song, Zhehua Zhou, Yuheng Huang, Da Song, Lei Ma",
+        "published": "2024-04-12",
+        "updated": "2024-04-12",
+        "primary_cat": "cs.SE",
+        "cats": [
+            "cs.SE",
+            "cs.AI",
+            "cs.CL",
+            "cs.CR",
+            "cs.LG"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.08472v1",
+        "title": "Selecting Shots for Demographic Fairness in Few-Shot Learning with Large Language Models",
+        "abstract": "Recently, work in NLP has shifted to few-shot (in-context) learning, with\nlarge language models (LLMs) performing well across a range of tasks. However,\nwhile fairness evaluations have become a standard for supervised methods,\nlittle is known about the fairness of LLMs as prediction systems. Further,\ncommon standard methods for fairness involve access to models weights or are\napplied during finetuning, which are not applicable in few-shot learning. Do\nLLMs exhibit prediction biases when used for standard NLP tasks? In this work,\nwe explore the effect of shots, which directly affect the performance of\nmodels, on the fairness of LLMs as NLP classification systems. We consider how\ndifferent shot selection strategies, both existing and new demographically\nsensitive methods, affect model fairness across three standard fairness\ndatasets. We discuss how future work can include LLM fairness evaluations.",
+        "authors": "Carlos Aguirre, Kuleen Sasse, Isabel Cachola, Mark Dredze",
+        "published": "2023-11-14",
+        "updated": "2023-11-14",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.02049v1",
+        "title": "Post Turing: Mapping the landscape of LLM Evaluation",
+        "abstract": "In the rapidly evolving landscape of Large Language Models (LLMs),\nintroduction of well-defined and standardized evaluation methodologies remains\na crucial challenge. This paper traces the historical trajectory of LLM\nevaluations, from the foundational questions posed by Alan Turing to the modern\nera of AI research. We categorize the evolution of LLMs into distinct periods,\neach characterized by its unique benchmarks and evaluation criteria. As LLMs\nincreasingly mimic human-like behaviors, traditional evaluation proxies, such\nas the Turing test, have become less reliable. We emphasize the pressing need\nfor a unified evaluation system, given the broader societal implications of\nthese models. Through an analysis of common evaluation methodologies, we\nadvocate for a qualitative shift in assessment approaches, underscoring the\nimportance of standardization and objective criteria. This work serves as a\ncall for the AI community to collaboratively address the challenges of LLM\nevaluation, ensuring their reliability, fairness, and societal benefit.",
+        "authors": "Alexey Tikhonov, Ivan P. Yamshchikov",
+        "published": "2023-11-03",
+        "updated": "2023-11-03",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "68T50",
+            "I.2.7"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2402.12150v1",
+        "title": "Your Large Language Model is Secretly a Fairness Proponent and You Should Prompt it Like One",
+        "abstract": "The widespread adoption of large language models (LLMs) underscores the\nurgent need to ensure their fairness. However, LLMs frequently present dominant\nviewpoints while ignoring alternative perspectives from minority parties,\nresulting in potential biases. We hypothesize that these fairness-violating\nbehaviors occur because LLMs express their viewpoints using a human personality\nthat represents the majority of training data. In response to this, we validate\nthat prompting LLMs with specific roles can allow LLMs to express diverse\nviewpoints. Building on this insight and observation, we develop FairThinking,\na pipeline designed to automatically generate roles that enable LLMs to\narticulate diverse perspectives for fair expressions. To evaluate FairThinking,\nwe create a dataset with a thousand items covering three fairness-related\ntopics and conduct experiments on GPT-3.5, GPT-4, Llama2, and Mistral to\ndemonstrate its superior performance.",
+        "authors": "Tianlin Li, Xiaoyu Zhang, Chao Du, Tianyu Pang, Qian Liu, Qing Guo, Chao Shen, Yang Liu",
+        "published": "2024-02-19",
+        "updated": "2024-02-19",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL",
+            "cs.AI",
+            "I.2; J.4"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2311.03033v1",
+        "title": "Beyond Words: A Mathematical Framework for Interpreting Large Language Models",
+        "abstract": "Large language models (LLMs) are powerful AI tools that can generate and\ncomprehend natural language text and other complex information. However, the\nfield lacks a mathematical framework to systematically describe, compare and\nimprove LLMs. We propose Hex a framework that clarifies key terms and concepts\nin LLM research, such as hallucinations, alignment, self-verification and\nchain-of-thought reasoning. The Hex framework offers a precise and consistent\nway to characterize LLMs, identify their strengths and weaknesses, and\nintegrate new findings. Using Hex, we differentiate chain-of-thought reasoning\nfrom chain-of-thought prompting and establish the conditions under which they\nare equivalent. This distinction clarifies the basic assumptions behind\nchain-of-thought prompting and its implications for methods that use it, such\nas self-verification and prompt programming.\n  Our goal is to provide a formal framework for LLMs that can help both\nresearchers and practitioners explore new possibilities for generative AI. We\ndo not claim to have a definitive solution, but rather a tool for opening up\nnew research avenues. We argue that our formal definitions and results are\ncrucial for advancing the discussion on how to build generative AI systems that\nare safe, reliable, fair and robust, especially in domains like healthcare and\nsoftware engineering.",
+        "authors": "Javier Gonz\u00e1lez, Aditya V. Nori",
+        "published": "2023-11-06",
+        "updated": "2023-11-06",
+        "primary_cat": "cs.LG",
+        "cats": [
+            "cs.LG",
+            "cs.AI"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2404.12736v1",
+        "title": "Large Language Model Supply Chain: A Research Agenda",
+        "abstract": "The rapid advancements in pre-trained Large Language Models (LLMs) and Large\nMultimodal Models (LMMs) have ushered in a new era of intelligent applications,\ntransforming fields ranging from natural language processing to content\ngeneration. The LLM supply chain represents a crucial aspect of the\ncontemporary artificial intelligence landscape. It encompasses the entire\nlifecycle of pre-trained models, from its initial development and training to\nits final deployment and application in various domains. This paper presents a\ncomprehensive overview of the LLM supply chain, highlighting its three core\nelements: 1) the model infrastructure, encompassing datasets and toolchain for\ntraining, optimization, and deployment; 2) the model lifecycle, covering\ntraining, testing, releasing, and ongoing maintenance; and 3) the downstream\napplication ecosystem, enabling the integration of pre-trained models into a\nwide range of intelligent applications. However, this rapidly evolving field\nfaces numerous challenges across these key components, including data privacy\nand security, model interpretability and fairness, infrastructure scalability,\nand regulatory compliance. Addressing these challenges is essential for\nharnessing the full potential of LLMs and ensuring their ethical and\nresponsible use. This paper provides a future research agenda for the LLM\nsupply chain, aiming at driving the continued advancement and responsible\ndeployment of these transformative LLMs.",
+        "authors": "Shenao Wang, Yanjie Zhao, Xinyi Hou, Haoyu Wang",
+        "published": "2024-04-19",
+        "updated": "2024-04-19",
+        "primary_cat": "cs.SE",
+        "cats": [
+            "cs.SE"
+        ],
+        "category": "LLM Fairness"
+    },
+    {
+        "url": "http://arxiv.org/abs/2401.08495v2",
+        "title": "Large Language Models Portray Socially Subordinate Groups as More Homogeneous, Consistent with a Bias Observed in Humans",
+        "abstract": "Large language models (LLMs) are becoming pervasive in everyday life, yet\ntheir propensity to reproduce biases inherited from training data remains a\npressing concern. Prior investigations into bias in LLMs have focused on the\nassociation of social groups with stereotypical attributes. However, this is\nonly one form of human bias such systems may reproduce. We investigate a new\nform of bias in LLMs that resembles a social psychological phenomenon where\nsocially subordinate groups are perceived as more homogeneous than socially\ndominant groups. We had ChatGPT, a state-of-the-art LLM, generate texts about\nintersectional group identities and compared those texts on measures of\nhomogeneity. We consistently found that ChatGPT portrayed African, Asian, and\nHispanic Americans as more homogeneous than White Americans, indicating that\nthe model described racial minority groups with a narrower range of human\nexperience. ChatGPT also portrayed women as more homogeneous than men, but\nthese differences were small. Finally, we found that the effect of gender\ndiffered across racial/ethnic groups such that the effect of gender was\nconsistent within African and Hispanic Americans but not within Asian and White\nAmericans. We argue that the tendency of LLMs to describe groups as less\ndiverse risks perpetuating stereotypes and discriminatory behavior.",
+        "authors": "Messi H. J. Lee, Jacob M. Montgomery, Calvin K. Lai",
+        "published": "2024-01-16",
+        "updated": "2024-04-26",
+        "primary_cat": "cs.CL",
+        "cats": [
+            "cs.CL"
+        ],
+        "category": "LLM Fairness"
+    }
+]
\ No newline at end of file