diff --git "a/related_34K/test_related_short_2405.01591v1.json" "b/related_34K/test_related_short_2405.01591v1.json" new file mode 100644--- /dev/null +++ "b/related_34K/test_related_short_2405.01591v1.json" @@ -0,0 +1,1438 @@ +[ + { + "url": "http://arxiv.org/abs/2405.01591v1", + "title": "Simplifying Multimodality: Unimodal Approach to Multimodal Challenges in Radiology with General-Domain Large Language Model", + "abstract": "Recent advancements in Large Multimodal Models (LMMs) have attracted interest\nin their generalization capability with only a few samples in the prompt. This\nprogress is particularly relevant to the medical domain, where the quality and\nsensitivity of data pose unique challenges for model training and application.\nHowever, the dependency on high-quality data for effective in-context learning\nraises questions about the feasibility of these models when encountering with\nthe inevitable variations and errors inherent in real-world medical data. In\nthis paper, we introduce MID-M, a novel framework that leverages the in-context\nlearning capabilities of a general-domain Large Language Model (LLM) to process\nmultimodal data via image descriptions. MID-M achieves a comparable or superior\nperformance to task-specific fine-tuned LMMs and other general-domain ones,\nwithout the extensive domain-specific training or pre-training on multimodal\ndata, with significantly fewer parameters. This highlights the potential of\nleveraging general-domain LLMs for domain-specific tasks and offers a\nsustainable and cost-effective alternative to traditional LMM developments.\nMoreover, the robustness of MID-M against data quality issues demonstrates its\npractical utility in real-world medical domain applications.", + "authors": "Seonhee Cho, Choonghan Kim, Jiho Lee, Chetan Chilkunda, Sujin Choi, Joo Heung Yoon", + "published": "2024-04-29", + "updated": "2024-04-29", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL", + "cs.AI", + "eess.IV" + ], + "label": "Original Paper", + "paper_cat": "LLM Fairness", + "gt": "In this section, we review the relevant literature on large language or multimodal models in the general domain as well as those applied specifically to the tasks in healthcare. 2.1. Large Language Models for Vision and Language Tasks Multimodal pretraining Large multimodal models have often leveraged language models as the backbone architecture and incorporate methods to connect the image encoder with the language model (Awadalla et al., 2023; Alayrac et al., 2022; Li et al., 2023a,b; Xu et al., 2024; Lauren\u00b8 con et al., 2024; Liu et al., 2024; Dai et al., 2024). Notably, LLaVA (Liu 1. https://commonfund.nih.gov/bridge2ai 3 Unimodal Approach to Multimodal Challenges in Radiology et al., 2024) employs a direct projection layer, BLIP-2 (Li et al., 2023b) uses a lightweight Q-Former, and Flamingo (Alayrac et al., 2022) uses gated cross-attention, to name a few examples. However, this paradigm suffers from two critical drawbacks. First, pretraining requires a high computational burden. Jointly aligning vision backbones and large language models requires large computational resources, despite considerably fewer trainable parameters compared to the total parameters. For example, Flamingo takes two weeks with 1,536 TPUs for the largest run, and BLIP-2 takes around 10 days with 16 A100 GPUs. Hence, it becomes prohibitively expensive to switch to a different large language model. The second limitation is the lack of modularity and flexibility. Even though they integrate pretrained image encoder and frozen LLM, updating this integrated system demands training the network with substantial data instances and time (Lauren\u00b8 con et al., 2024), hindering iterative research and development. Multimodal Reasoning with Language There have also been approaches that attempt to tackle multimodal tasks using language models alone (Yang et al., 2022; Guo et al., 2022; Hu et al., 2022; Xie et al., 2022; Sur\u00b4 \u0131s et al., 2023; Berrios et al., 2023). Among these, LENS (Berrios et al., 2023) is particularly relevant to our work, as this framework extracts generic information such as tags, attributes, and captions with three different vision modules, and provides them to the language model. While this method can be beneficial for general domain tasks, such an approach does not scale well to tasks that require expert knowledge. In contrast, our framework facilitates LLM\u2019s multimodal reasoning within the medical domain. We introduce a more streamlined framework with a single \u2018vision module\u2019 and tailor a general domain LLM specifically for medical tasks, by leveraging the adaptability of the LLM in this knowledge-intensive domain. 2.2. Large Language Models for Healthcare Domain Several efforts have been made to specialize these general-purpose foundation models through fine-tuning on healthcare domain-specific data (Li et al., 2024; Moor et al., 2023; Zhang et al., 2023a; Wu et al.; Tu et al., 2024; Sun et al., 2023). However, task-specific fine-tuning is often inefficient, requiring large amounts of high-quality medical data and extensive computing resources. Consequently, efficient adaptation methods are also explored, which primarily follow two categories: lightweight fine-tuning techniques (Hu et al., 2021; Houlsby et al., 2019), and leveraging the models\u2019 in-context learning capabilities in a few-shot setting. While finetuned models generally outperform their counterparts in few-shot setting(Van Veen et al., 2023b,a; Lewis et al., 2020), there is a growing interest in exploring domain adaptation techniques, mainly in-context learning, which do not require additional training (Agrawal et al., 2022; Li\u00b4 evin et al., 2023; Byra et al., 2023; Yan et al., 2023; Nori et al., 2023; Van Veen et al., 2023a). In-context learning has many potential advantages for the clinical domain because there is often a limited set of labeled data due to the high level of expertise needed for annotation. Our work also aligns with this in-context learning approach. By leveraging the capabilities of general-purpose LLMs without additional fine-tuning, our approach aims to overcome the limitations of task-specific fine-tuning while maintaining the models\u2019 broad knowledge and generalization abilities. 4 Unimodal Approach to Multimodal Challenges in Radiology", + "pre_questions": [], + "main_content": "Introduction Recently, large multimodal models (LMMs) have made significant progress, becoming capable of interpreting and integrating information across different data types, such as text and images (Achiam et al., 2023; Driess et al., 2023). This progress closely aligns to advancements in Large Language Models (LLMs), since many LMMs are built upon the structural foundation of pre-trained LLMs (Touvron et al., 2023; Chiang et al., 2023). Consequently, attempts to use LMM\u2019s in-context learning capability without fine-tuning on specific tasks are increasing (Wei et al., 2022; Zhang et al., 2023b). This tendency also appears in the medical domain, where the data collection is challenging due to the concerns on using sensitive clinical information containing re-identifiable patient data (Benitez and Malin, 2010) and the expensive labeling process which requires expert knowledge. However, the performance of in-context learning approach is highly susceptible to the data quality (Liu et al., 2023). One primary cause of degrading data quality is data loss occurring during the data collection and curation process, even when guided by expert input. For instance, the analysis of radiographic images has reported an error rate of 3 to 5% (Robinson et al., 1999) and the problem appears challenging to resolve until now (Brady, 2017). Similarly, a basic chart review showed electronic health record (EHR) errors could reach 9 to 10% (Feng et al., 2020). Additionally, variations in data interpretation by medical professionals are quite common (Whiting et al., 2015) and this can further exacerbate these issues by adding another layer of complexity to ensuring the quality and consistency of data. In this paper, we introduce MID-M, a Multimodal framework with Image Description for Medical domain. It is an in-context learning framework that demonstrates robust performance even with low-quality data. Our framework uses general-domain LLM and leverages multimodal data by converting images to textual descriptions. Notably, it achieves comparable performance to other general-domain and fine-tuned LMMs without pre-training on multimodality and extensive fine-tuning for the medical domain. In addition, by processing the image as a text description, it has the advantage of representing the image in an accessible and interpretable way, compared to traditional embedded vector representations. Our framework is illustrated in Figure 1. To evaluate the model\u2019s performance in the scenario with low-quality data, we systematically mask the texts in a medical dataset and make the new adversarial evaluation data. We compare the performance of general domain pre-trained and medically fine-tuned LMM models to our framework in a few-shot setting, focusing on their accuracy and semantic comprehensibility. Through this experiment, we aim to assess the effectiveness of LMMs in generalizing to domain-specific tasks and managing incomplete source data. 2 Unimodal Approach to Multimodal Challenges in Radiology Figure 1: Overview of MID-M framework. Generalizable Insights about Machine Learning in the Context of Healthcare In this study, we analyze the generalization capabilities of LLMs and LMMs across various dimensions. First, we investigate the ability of models trained in general domains to effectively generalize to medical texts through few-shot learning. This aligns with the recent trend towards adopting eco-friendly approaches in AI model training by reducing the need for massive pretraining and extensive fine-tuning. It also contributes to the validation of model\u2019s potential to generalize well across any domain or task, as aimed for in general domain research. Second, we verify the models\u2019 ability to generalize in situations where data quality deteriorates. The challenge that in-context learning models are easily influenced by the number of few-shot samples and the quality of trained data remains unresolved. Therefore, concrete evaluation in this regard is crucial for the advancement of models. Furthermore, our research reveals that text-only models can efficiently handle multimodal healthcare tasks with less computational demand. We demonstrate that smaller language models can achieve results comparable to larger models, indicating a path for wider adoption of AI in healthcare, especially in resource-limited settings in computation or a lower level of healthcare environment. This goes in line with a move towards developing globally accessible AI solutions by the National Institutes of Health1, highlighting the potential utility of proposed research in expanding AI\u2019s reach. This section introduces MID-M, an efficient in-context learning framework that performs multimodal tasks using only small-sized language models. MID-M consists of two key components: a domain-specific image converter and a general-domain language model. Image Conversion with Domain-Specific Image Classifier To reformulate visionlanguage tasks to text-to-text task, we first extract key information from images and convert it to text. Taking inspiration from Classification-via-Description (Menon and Vondrick, 2022) which showed good generalization performance, we employ a Description-viaClassification strategy. Classification-via-description involves generating textual features representing each class with a language model, then analyzing whether the input exhibits those features to make the final classification decision. Conversely, we use the classification result to generate an image description. Note that using a general-domain image module trained only on generic data, as in Berrios et al. (2023), cannot yield meaningful classification results since chest X-rays require domain knowledge. Therefore, we utilize a SOTA model (Chong et al., 2023) for CheXpert, a publicly available large chest X-ray classification dataset. This classifer computes a probability of 14 major diseases for chest radiology images. For diseases that the classifier predicts to have a probability of presence greater than 20%, we use the sentence \u201cIt seems there is {disease} in the image.\u201d Conversely, for diseases predicted as absent, we use \u201cIt seems there is no {disease} in the image.\u201d These sentences describing the presence or absence of each disease are all concatenated and used as an image description. This method can be extended to other domains without domain expertise, by simply replacing the classifier with a domain-specific one. We provide an example of our prompt in Appendix A. General-domain LLM We utilize a general-domain language model for multimodal reasoning. While chains-of-thought (Wei et al., 2022) are generally advantageous for domain adaptation and complex reasoning, their benefits are not consistent across generation tasks (Kim et al., 2023). Furthermore, it requires domain knowledge to generate rationales. Thus, we utilize strategies known to be helpful for domain adaptation (Van Veen et al., 2023a) instead. Specifically, we prepend a sentence assigning the role of an AI assistant as a domain expert: \u201cYou are an expert medical professional.\u201d This is followed by a detailed task description crafted for the radiology domain. For the few-shot examples, we provide the image description, corresponding finding, and impression in sequence. During inference, we omit the impression from the examples to be completed by the model. We use a BM25 retriever, known for its capability to compute token overlap quickly and retrieve relevant examples effectively. 4. Dataset We utilized the MIMIC-CXR dataset (Johnson et al., 2019), a de-identified and publicly available chest radiograph database collected from 2011 to 2016. It has a total number of 227,835 images and 128,032 paired reports from 65,379 patients. Paired reports include finding and impression. A finding is a detailed observation made from the X-ray images written by radiologists, and an impression is a concise summary of the finding which becomes the primary mode of communication between medical professionals. 5 Unimodal Approach to Multimodal Challenges in Radiology Table 1: Example of MIMIC-CXR findings with different masking rate. Full Corrupted 0.1 Corrupted 0.3 Corrupted 0.5 PA and lateral views of the chest provided. AICD projects over the left chest wall with lead tip extending to the region of the right ventricle. The heart is mildly enlarged. There is no evidence of pneumonia or CHF. No effusion or pneumothorax seen. Bony structures are intact. PA and lateral views the chest provided. AICD projects over the chest wall with lead tip extending to the region of the right ventric . heart is mildly enlarged. there is no of pneumonia or CHF. effusion or pneumothorax seen. Bony structures are intact. PA and lateral of chest provided. CD projects over the left wall with lead tip to the region of the ventric the heart is . of pneumonia or CH . no effusion or pneum rax seen. intact. PA and lateral views of provided. AI the chest with tip to the vent heart is enlarged. there no F no usion or pneum rax structures are intact Preliminary Analysis Our examination of the MIMIC-CXR dataset revealed 3,146 samples (2.46% of the total) that were unsuitable for research due to one or more of the following reasons: the finding label contained fewer than three words, was shorter than the impression label, or had more than three words masked. This level of incompleteness aligns with error rates reported in previous studies (Robinson et al., 1999). Pre-preprocessing To prepare the data for our study, we first merged the original training, validation, and test sets. We then removed samples with the findings at both extremes of length, removing the shortest and longest 25% of findings to ensure a meaningful level of detail remained after applying masking techniques. This process resulted in a dataset of 64,613 images and reports, which we then randomly split into training, validation, and test sets containing 62,613, 1,000, and 1,000 samples, respectively. Masking To further evaluate the robustness of LLMs against incomplete data, we employed a masking technique akin to that used by BERT (Devlin et al., 2018), randomly deleting words or phrases at subword levels. We conducted masking at three different rates to the original test set and yielded four distinct test sets: original (full), corrupted at 10% (corrupted 0.1), corrupted at 30% (corrupted 0.3), and corrupted at 50% (corrupted 0.5). We evaluated the effectiveness of our corruption strategy using the F1CheXbert (Zhang et al., 2020) metric. CheXbert (Smit et al., 2020) is designed to identify 14 medical observations in chest X-ray images. The F1CheXbert score is calculated by comparing CheXbert\u2019s predictions on the generated text against the corresponding reference text. We compare the findings from full against the findings from each corrupted test set. We discovered a strong correlation between the extent of text corruption and the precision of disease mentions. The micro-average F1CheXbert scores were 91.2% for the 10% corrupted set, 69.3% for the 30% corrupted set, and 51.1% for the 50% corrupted set. Examples of corrupted text are illustrated in Table 1. 6 Unimodal Approach to Multimodal Challenges in Radiology 5. Experimental Setup 5.1. Task In this study, the task requires the model to generate an impression based on an X-ray image and its corresponding findings. For all experiments, we employed a 2-shot learning approach, utilizing the BM25 algorithm to retrieve two training samples with findings similar to the findings of the test sample. When experimenting with corrupted test sets, the samples are retrieved based on the corrupted findings. This task is designed to comprehensively evaluate the model\u2019s ability to understand medical images and clinical data, as well as to identify salient information even when incomplete data is given. 5.2. Baselines We have two groups of baselines in our experiments: LMMs that are pre-trained in the general domain and those further fine-tuned with medical domain data. However, considering the constraints of a few-shot learning setting, we found a limited number of LMMs are capable of processing multiple independent images in a single prompt. Notably, Flamingo (Alayrac et al., 2022), is trained on interleaved text and image pairs and supports multiple images. We used its open-sourced version, OpenFlamingo (Awadalla et al., 2023). This functionality is also applied to the models that are based on Flamingo, IDEFICS (Lauren\u00b8 con et al., 2024) and OTTER (Li et al., 2023a). For models trained on data from the medical domain, we selected MedFlamingo (Moor et al., 2023) and RadFM (Wu et al., 2023). MedFlamingo is trained with medical publications and textbooks and is based on OpenFlamingo. RadFM is a specialized model for the radiology domain and pre-trained with large-scale, high-quality multimodal dataset named MedMD. It\u2019s important to note that although MedMD includes MIMIC-CXR, we still chose to include it in our baseline set. Despite data leakage concerns in our full test setting, we believe that evaluating the model\u2019s performance on corrupted versions of MIMIC-CXR would provide valuable insights into its robustness and generalization capabilities. A detailed comparison is presented in Table 2. Note that all the baseline models have 9 billion parameters or more, whereas our backbone model, Flan-T5-xl, has only 3 billion parameters. For the baselines, we adhered to the example code in their public GitHub repositories when available; otherwise, we utilized a consistent prompt with the example code from other baselines. 5.3. Evaluation The performance of the models are compared using two different LLM evaluation methods, ROUGE-L (Lin, 2004) and F1RadGraph (Delbrouck et al., 2022). ROUGE, the RecallOriented Understudy for Gisting Evaluation, is a metric widely used in the text summarization task. This score evaluates the similarity between the generated text and the reference text based on token overlap. Specifically, we used ROUGE-L which measures the longest common subsequence (LCS) to assess similarity at the sentence level. F1RadGraph is a metric that evaluates medical entities and relations within the generated reports. It utilizes a PubMedBERT (Gu et al., 2021) fine-tuned on the RadGraph (Jain et al., 2021), a dataset with chest X-ray radiology reports annotated with medical entities and their relations. This 7 Unimodal Approach to Multimodal Challenges in Radiology Table 2: Descriptions of baseline models. The first four models are general-domain LMMs that can handle multiple images in one input. The two models in the middle are LMMs that are fine-tuned with medical domain data. Model Size Domain Image Encoder Language Model OtterM 9B General ViT-L/14 MPT-7b OtterL 9B General ViT-L/14 LLaMA-7b OpenFlamingo 9B General ViT-L/14 MPT-7b IDEFICS 9B General ViT-H/14 LLaMA-7b MedFlamingo 9B Medical ViT-L/14 LLaMA-7b RadFM 14B Radiology ViT-B/32 MedLLaMA-13b MID-M 3B General Classifier(7M) Flan-T5-xl fine-tuned PubMedBERT model identifies the entities and relations from the generated reports and the reference reports. When the generated report correctly matches the entities and their relation to that in the reference, they receive a reward of 1, and otherwise, 0. The final F1 score is computed based on this reward score. Both metrics are implemented using the Python Package Index (PyPI)2 6. Experiment 6.1. Result The experimental results are presented in Table 3. MID-M achieves performance comparable to that of other baseline models in all settings, despite using only a third of the parameters. It even surpasses other models in experiments with corrupted data, highlighting its exceptional generalization capabilities. RadFM, which previously trained on MIMIC-CXR, achieves the highest score on the full test set, which is not a surprising result. IDEFICS and OpenFlamingo demonstrate strong generalization ability, achieving scores around 0.4 across both ROUGE and F1RadGraph metrics. However, we observed that some models are highly sensitive to even a minor level of data corruption (i.e. 0.1). For instance, RadFM suffers from a significant performance drop in corrupted settings and shows performance similar to or even worse than large-scale general domain models. This sensitivity to text corruption is likely due to its training on only high-quality radiology images and associated texts. It is critical problem in clinical applications where even small decreases in performance can significantly impact patient care. Otter and OpenFlamingo also shows considerable performance declines in corrupted settings. Conversely, IDEFICS, MedFlamingo and our framework maintain robust performance on low-quality data. As the text masking probability increases (i.e. 0.3, 0.5), our model begins to demonstrate its full potential. It achieves the highest scores under the ROUGE metric and exhibits either comparable or superior performance on the F1RadGraph metric. 2. https://pypi.org/project/radgraph/ 8 Unimodal Approach to Multimodal Challenges in Radiology Table 3: Model performances in each test set with regard to the accuracy. The highest score is bolded and the second highest is underlined. Method Full Corrupted 0.1 Corrupted 0.3 Corrupted 0.5 ROUGE OtterM 0.3140 0.2258 0.2116 0.1894 OtterL 0.2655 0.2089 0.1876 0.1461 OpenFlamingo 0.4310 0.3458 0.3420 0.2888 IDEFICS 0.3951 0.3776 0.3321 0.2807 MedFlamingo 0.3804 0.3622 0.3270 0.2784 RadFM 0.5222 0.3952 0.3299 0.2534 MID-M(ours) 0.4064 0.3719 0.3438 0.2977 F1RadGraph OtterM 0.2801 0.2114 0.1704 0.1327 OtterL 0.2587 0.1994 0.1720 0.1237 OpenFlamingo 0.4030 0.3427 0.3151 0.2541 IDEFICS 0.4089 0.3688 0.3064 0.2543 MedFlamingo 0.3973 0.3510 0.3046 0.2566 RadFM 0.5054 0.3839 0.3094 0.2363 MID-M(ours) 0.3732 0.3379 0.2973 0.2622 It is also surprising that MedFlamingo, which is further fine-tuned from OpenFlamingo with medical domain text, does not outperform OpenFlamingo. This may be because MedFlamingo is primarily fine-tuned for medical visual question-answering tasks rather than various medical generation tasks. These experimental results shows the importance of task setting during pre-training and fine-tuning. 6.2. Ablation Studies To comprehend the the effect of each component in our framework to the overall performance, we systematically remove each element textual input (findings), image input (descriptions), and both. In scenarios where both text and images are excluded, the model must generate impressions based solely on two impressions retrieved based on the similarity of findings. The results of these ablation studies are presented in Table 4. The results indicate that the multimodal framework exhibits the strongest generalization capability by leveraging all available information. Since the impressions are primarily derived from findings, the text component alone proves to be most effective when the text data encompasses sufficient information. Still, using both image and text information achieves comparable scores and outperforms in the corrupted test sets. When text corruption exceeds a 10% probability, incorporating image descriptions improves the model\u2019s accuracy compared to using only the findings or only the image descriptions in most cases. One interesting point is, at the highest level of text corruption, the corrupted text acts as noise 9 Unimodal Approach to Multimodal Challenges in Radiology Table 4: Model performances in different input settings. Method Full Corrupted 0.1 Corrupted 0.3 Corrupted 0.5 ROUGE MID-M 0.4064 0.3719 0.3438 0.2977 w/o text 0.3794 0.3563 0.3440 0.2942 w/o image 0.4119 0.3683 0.3218 0.2550 w/o text and image 0.3649 0.3309 0.3129 0.2670 F1RadGraph MID-M 0.3732 0.3379 0.2973 0.2622 w/o text 0.3354 0.3185 0.2948 0.2575 w/o image 0.3918 0.3323 0.2784 0.2247 w/o text and image 0.3298 0.2976 0.2783 0.2378 and severely impairs the model\u2019s performance, leading to worse outcomes than when both text and images were excluded. This trend remains consistent across evaluation metrics of token overlap, measured by ROUGE, and semantic coherence, measured by F1RadGraph. 6.3. Disease Identification Ability In this section, we further analyze the model\u2019s ability to identify the major diseases in X-rays and corresponding findings. The evaluation in Section 6.1 is based on a broad comparison of impressions and predictions at the word and semantic level. Additionally in this section, we use the F1CheXbert to assess whether the model identifies all the key diseases noted in the gold impressions in its predictions as well. We report both the micro-average F1 score and the individual F1 scores for five main diseases, as shown in Table 5. Table 5: Comparison of MID-M\u2019s performance on the identification of major diseases under different levels of data corruption. Abbreviations: CMG, cardiomegaly; Consol, consolidation; Atelect, atelectasis; PE, pleural effusion. CMG Edema Consol Atelect PE Micro Avg. Full 0.4200 0.5714 0.4444 0.4916 0.6335 0.5302 Corrupted 0.1 0.3204 0.4923 0.1538 0.4067 0.5543 0.4379 Corrupted 0.3 0.3010 0.3884 0.1176 0.3972 0.4250 0.3724 Corrupted 0.5 0.2631 0.3606 0.1538 0.2704 0.4467 0.3327 Support 87 148 25 151 138 549 We found that the model\u2019s performance declines with increased data corruption, but the rate of decline varies across diseases. For instance, the model\u2019s ability to identify pleural effusion (PE) remains relatively robust against corruption, even under conditions of high data corruption, compared to other diseases. In contrast, our model experiences a significant drop in consolidation (Consol) even at the lowest level of corruption (0.1). This suggests that the model manages to recognize some diseases in a manner less susceptible 10 Unimodal Approach to Multimodal Challenges in Radiology Table 6: Model performances in different number of shots in the input prompt. Method Full Corrupted 0.1 Corrupted 0.3 Corrupted 0.5 ROUGE 0-shot 0.1673 0.1270 0.1127 0.1091 1-shot 0.3818 0.3438 0.3172 0.2180 OURS (2-shot) 0.4064 0.3719 0.3438 0.2977 F1RadGraph 0-shot 0.1573 0.0994 0.072 0.0439 1-shot 0.3536 0.3091 0.2797 0.2386 OURS (2-shot) 0.3732 0.3379 0.2973 0.2622 to corruption. This might be attributed to the feasibility of retrieving similar samples. When there are many samples with a certain disease in the training set, it becomes easier to retrieve relevant samples for the diseases and vice versa. In addition, a significant drop in consolidation might be attributed to a unique linguistic challenge. When some of the subwords from consolidation are obscured, the remaining segments (such as \u2019solid\u2019 or \u2019tion\u2019) could be mistakenly associated with other terms frequently used in radiology reports, complicating its identification. 6.4. In-Context Learning Ability In this section, we evaluate the model\u2019s in-context learning capabilities by varying the number of examples (shots) included in the prompt. Previous experiments, as discussed in Section 6.1, uniformly used a two-shot setting. In addition to this, we explore the model\u2019s abilities in zero-shot and one-shot settings. The results are presented in Table 6. We observe a decline in performance in both zero-shot and one-shot settings compared to the two-shot scenario. However, the disparity in performance across these settings is noteworthy. In the zero-shot setting, the model struggles significantly, achieving a ROUGE score of only about 0.1673 on the full test set and with the F1RadGraph score approaching zero across all corrupted test sets. In contrast, introducing just one relevant example enhances the model\u2019s performance substantially, achieving results that are comparatively close to those in the two-shot setting in many cases. This underscores the impressive incontext learning capabilities of pretrained large models, highlighting their potential for easy adaptation to different domains without the need for fine-tuning. 7. Discussion In this paper, we introduce MID-M, a novel multimodal framework designed for the medical domain, which leverages the in-context learning capabilities of a general-domain LLM. By transforming images into textual descriptions, MID-M facilitates an interpretable representation of medical images, enabling complex information to be more accessible to engineers and, potentially, to practitioners. Notably, MID-M demonstrates superior generalization capability, outperforming models with substantially more parameters that are extensively 11 Unimodal Approach to Multimodal Challenges in Radiology trained for multimodality and medical applications. This emphasizes the potential of leveraging general-domain models for specialized tasks, offering a sustainable and cost-effective approach without massive pretraining and extensive fine-tuning. This approach aligns with the ongoing shift towards developing powerful AI tools in healthcare. Furthermore, it highlights the importance of robustness and generalization in systems, particularly in healthcare settings where data quality can vary significantly. The framework\u2019s ability to maintain robust performance even with degraded data quality, presents a compelling case for its application in real-world medical scenarios even with limited resources. We hope our work can contribute to the development of globally accessible AI systems. Limitations Despite the contributions of our research, there are a few limitations to our work. First, while MID-M demonstrates superior in-context learning capability, it still underperforms compared to task-specific fine-tuned models. This is observed from the performance of RadFM on the full test set. However, it is important to note that models like RadFM demand high computational resources for their training and show a significant drop in generalization ability when faced with incomplete data. In future works, we could focus on developing in-context learning approaches that can achieve performance comparable to that of task-specific models, even under ideal conditions. The second limitation is the artificial nature of our masking approach. In real-world applications, masking is likely to occur at the word level and targets sensitive information. Our method, however, employed subword-level random masking. Despite this fundamental difference, we believe our test set includes samples with realistically plausible masking since sub-word masking cumulatively approximates word-level masking, and sensitive information can also be obscured by randomly chosen. Our statistical analysis, which shows main diseases are masked in proportion to the corruption rate, also supports our hypothesis. Moreover, this approach allows for the systematic masking with which we produced four different test sets. We believe the insights derived from our experiments can provide valuable guidance for future research. Acknowledgments This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government(MSIT) (RS-2022-00143911,AI Excellence Global Innovative Leader Education Program). Yoon JH has received grant support from NIH K23GM138984." + }, + { + "url": "http://arxiv.org/abs/2305.01146v3", + "title": "RadAdapt: Radiology Report Summarization via Lightweight Domain Adaptation of Large Language Models", + "abstract": "We systematically investigate lightweight strategies to adapt large language\nmodels (LLMs) for the task of radiology report summarization (RRS).\nSpecifically, we focus on domain adaptation via pretraining (on natural\nlanguage, biomedical text, or clinical text) and via discrete prompting or\nparameter-efficient fine-tuning. Our results consistently achieve best\nperformance by maximally adapting to the task via pretraining on clinical text\nand fine-tuning on RRS examples. Importantly, this method fine-tunes a mere\n0.32% of parameters throughout the model, in contrast to end-to-end fine-tuning\n(100% of parameters). Additionally, we study the effect of in-context examples\nand out-of-distribution (OOD) training before concluding with a radiologist\nreader study and qualitative analysis. Our findings highlight the importance of\ndomain adaptation in RRS and provide valuable insights toward developing\neffective natural language processing solutions for clinical tasks.", + "authors": "Dave Van Veen, Cara Van Uden, Maayane Attias, Anuj Pareek, Christian Bluethgen, Malgorzata Polacin, Wah Chiu, Jean-Benoit Delbrouck, Juan Manuel Zambrano Chaves, Curtis P. Langlotz, Akshay S. Chaudhari, John Pauly", + "published": "2023-05-02", + "updated": "2023-07-20", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2212.10846v3", + "title": "From Images to Textual Prompts: Zero-shot VQA with Frozen Large Language Models", + "abstract": "Large language models (LLMs) have demonstrated excellent zero-shot\ngeneralization to new language tasks. However, effective utilization of LLMs\nfor zero-shot visual question-answering (VQA) remains challenging, primarily\ndue to the modality disconnection and task disconnection between LLM and VQA\ntask. End-to-end training on vision and language data may bridge the\ndisconnections, but is inflexible and computationally expensive. To address\nthis issue, we propose \\emph{Img2Prompt}, a plug-and-play module that provides\nthe prompts that can bridge the aforementioned modality and task\ndisconnections, so that LLMs can perform zero-shot VQA tasks without end-to-end\ntraining. In order to provide such prompts, we further employ LLM-agnostic\nmodels to provide prompts that can describe image content and self-constructed\nquestion-answer pairs, which can effectively guide LLM to perform zero-shot VQA\ntasks. Img2Prompt offers the following benefits: 1) It can flexibly work with\nvarious LLMs to perform VQA. 2)~Without the needing of end-to-end training, it\nsignificantly reduces the cost of deploying LLM for zero-shot VQA tasks. 3) It\nachieves comparable or better performance than methods relying on end-to-end\ntraining. For example, we outperform Flamingo \\cite{Deepmind:Flamingo2022} by\n5.6\\% on VQAv2. On the challenging A-OKVQA dataset, our method even outperforms\nfew-shot methods by as much as 20\\%.", + "authors": "Jiaxian Guo, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Boyang Li, Dacheng Tao, Steven C. H. Hoi", + "published": "2022-12-21", + "updated": "2023-05-08", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.MM" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2305.17100v2", + "title": "BiomedGPT: A Unified and Generalist Biomedical Generative Pre-trained Transformer for Vision, Language, and Multimodal Tasks", + "abstract": "Conventional task- and modality-specific artificial intelligence (AI) models\nare inflexible in real-world deployment and maintenance for biomedicine. At the\nsame time, the growing availability of biomedical data, coupled with the\nadvancements in modern multi-modal multi-task AI techniques, has paved the way\nfor the emergence of generalist biomedical AI solutions. These solutions hold\nthe potential to interpret different medical modalities and produce expressive\noutputs such as free-text reports or disease diagnosis. Here, we propose\nBiomedGPT, the first open-source and generalist visual language AI for diverse\nbiomedical tasks. BiomedGPT achieved 16 state-of-the-art results across five\nclinically significant tasks on 26 datasets. Notably, it outperformed OpenAI's\nGPT-4 with vision (GPT-4V) in radiology human evaluation and surpassed Google's\nMed-PaLM M (12B) in breast cancer diagnosis and medical visual question\nanswering. Moreover, BiomedGPT facilitates zero-shot transfer learning, greatly\nenhancing its utility as a biomedical assistant, similar to ChatGPT. Our method\ndemonstrates effective training with diverse datasets can lead to more\npractical biomedical AI.", + "authors": "Kai Zhang, Jun Yu, Eashan Adhikarla, Rong Zhou, Zhiling Yan, Yixin Liu, Zhengliang Liu, Lifang He, Brian Davison, Xiang Li, Hui Ren, Sunyang Fu, James Zou, Wei Liu, Jing Huang, Chen Chen, Yuyin Zhou, Tianming Liu, Xun Chen, Yong Chen, Quanzheng Li, Hongfang Liu, Lichao Sun", + "published": "2023-05-26", + "updated": "2024-01-09", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL", + "cs.AI" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2301.12597v3", + "title": "BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models", + "abstract": "The cost of vision-and-language pre-training has become increasingly\nprohibitive due to end-to-end training of large-scale models. This paper\nproposes BLIP-2, a generic and efficient pre-training strategy that bootstraps\nvision-language pre-training from off-the-shelf frozen pre-trained image\nencoders and frozen large language models. BLIP-2 bridges the modality gap with\na lightweight Querying Transformer, which is pre-trained in two stages. The\nfirst stage bootstraps vision-language representation learning from a frozen\nimage encoder. The second stage bootstraps vision-to-language generative\nlearning from a frozen language model. BLIP-2 achieves state-of-the-art\nperformance on various vision-language tasks, despite having significantly\nfewer trainable parameters than existing methods. For example, our model\noutperforms Flamingo80B by 8.7% on zero-shot VQAv2 with 54x fewer trainable\nparameters. We also demonstrate the model's emerging capabilities of zero-shot\nimage-to-text generation that can follow natural language instructions.", + "authors": "Junnan Li, Dongxu Li, Silvio Savarese, Steven Hoi", + "published": "2023-01-30", + "updated": "2023-06-15", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2306.00890v1", + "title": "LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day", + "abstract": "Conversational generative AI has demonstrated remarkable promise for\nempowering biomedical practitioners, but current investigations focus on\nunimodal text. Multimodal conversational AI has seen rapid progress by\nleveraging billions of image-text pairs from the public web, but such\ngeneral-domain vision-language models still lack sophistication in\nunderstanding and conversing about biomedical images. In this paper, we propose\na cost-efficient approach for training a vision-language conversational\nassistant that can answer open-ended research questions of biomedical images.\nThe key idea is to leverage a large-scale, broad-coverage biomedical\nfigure-caption dataset extracted from PubMed Central, use GPT-4 to\nself-instruct open-ended instruction-following data from the captions, and then\nfine-tune a large general-domain vision-language model using a novel curriculum\nlearning method. Specifically, the model first learns to align biomedical\nvocabulary using the figure-caption pairs as is, then learns to master\nopen-ended conversational semantics using GPT-4 generated instruction-following\ndata, broadly mimicking how a layperson gradually acquires biomedical\nknowledge. This enables us to train a Large Language and Vision Assistant for\nBioMedicine (LLaVA-Med) in less than 15 hours (with eight A100s). LLaVA-Med\nexhibits excellent multimodal conversational capability and can follow\nopen-ended instruction to assist with inquiries about a biomedical image. On\nthree standard biomedical visual question answering datasets, LLaVA-Med\noutperforms previous supervised state-of-the-art on certain metrics. To\nfacilitate biomedical multimodal research, we will release our\ninstruction-following data and the LLaVA-Med model.", + "authors": "Chunyuan Li, Cliff Wong, Sheng Zhang, Naoto Usuyama, Haotian Liu, Jianwei Yang, Tristan Naumann, Hoifung Poon, Jianfeng Gao", + "published": "2023-06-01", + "updated": "2023-06-01", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.CL" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2304.08485v2", + "title": "Visual Instruction Tuning", + "abstract": "Instruction tuning large language models (LLMs) using machine-generated\ninstruction-following data has improved zero-shot capabilities on new tasks,\nbut the idea is less explored in the multimodal field. In this paper, we\npresent the first attempt to use language-only GPT-4 to generate multimodal\nlanguage-image instruction-following data. By instruction tuning on such\ngenerated data, we introduce LLaVA: Large Language and Vision Assistant, an\nend-to-end trained large multimodal model that connects a vision encoder and\nLLM for general-purpose visual and language understanding.Our early experiments\nshow that LLaVA demonstrates impressive multimodel chat abilities, sometimes\nexhibiting the behaviors of multimodal GPT-4 on unseen images/instructions, and\nyields a 85.1% relative score compared with GPT-4 on a synthetic multimodal\ninstruction-following dataset. When fine-tuned on Science QA, the synergy of\nLLaVA and GPT-4 achieves a new state-of-the-art accuracy of 92.53%. We make\nGPT-4 generated visual instruction tuning data, our model and code base\npublicly available.", + "authors": "Haotian Liu, Chunyuan Li, Qingyang Wu, Yong Jae Lee", + "published": "2023-04-17", + "updated": "2023-12-11", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.AI", + "cs.CL", + "cs.LG" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2308.01390v2", + "title": "OpenFlamingo: An Open-Source Framework for Training Large Autoregressive Vision-Language Models", + "abstract": "We introduce OpenFlamingo, a family of autoregressive vision-language models\nranging from 3B to 9B parameters. OpenFlamingo is an ongoing effort to produce\nan open-source replication of DeepMind's Flamingo models. On seven\nvision-language datasets, OpenFlamingo models average between 80 - 89% of\ncorresponding Flamingo performance. This technical report describes our models,\ntraining data, hyperparameters, and evaluation suite. We share our models and\ncode at https://github.com/mlfoundations/open_flamingo.", + "authors": "Anas Awadalla, Irena Gao, Josh Gardner, Jack Hessel, Yusuf Hanafy, Wanrong Zhu, Kalyani Marathe, Yonatan Bitton, Samir Gadre, Shiori Sagawa, Jenia Jitsev, Simon Kornblith, Pang Wei Koh, Gabriel Ilharco, Mitchell Wortsman, Ludwig Schmidt", + "published": "2023-08-02", + "updated": "2023-08-07", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.AI", + "cs.LG" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2307.15189v1", + "title": "Med-Flamingo: a Multimodal Medical Few-shot Learner", + "abstract": "Medicine, by its nature, is a multifaceted domain that requires the synthesis\nof information across various modalities. Medical generative vision-language\nmodels (VLMs) make a first step in this direction and promise many exciting\nclinical applications. However, existing models typically have to be fine-tuned\non sizeable down-stream datasets, which poses a significant limitation as in\nmany medical applications data is scarce, necessitating models that are capable\nof learning from few examples in real-time. Here we propose Med-Flamingo, a\nmultimodal few-shot learner adapted to the medical domain. Based on\nOpenFlamingo-9B, we continue pre-training on paired and interleaved medical\nimage-text data from publications and textbooks. Med-Flamingo unlocks few-shot\ngenerative medical visual question answering (VQA) abilities, which we evaluate\non several datasets including a novel challenging open-ended VQA dataset of\nvisual USMLE-style problems. Furthermore, we conduct the first human evaluation\nfor generative medical VQA where physicians review the problems and blinded\ngenerations in an interactive app. Med-Flamingo improves performance in\ngenerative medical VQA by up to 20\\% in clinician's rating and firstly enables\nmultimodal medical few-shot adaptations, such as rationale generation. We\nrelease our model, code, and evaluation app under\nhttps://github.com/snap-stanford/med-flamingo.", + "authors": "Michael Moor, Qian Huang, Shirley Wu, Michihiro Yasunaga, Cyril Zakka, Yash Dalmia, Eduardo Pontes Reis, Pranav Rajpurkar, Jure Leskovec", + "published": "2023-07-27", + "updated": "2023-07-27", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.AI" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2206.01843v2", + "title": "Visual Clues: Bridging Vision and Language Foundations for Image Paragraph Captioning", + "abstract": "People say, \"A picture is worth a thousand words\". Then how can we get the\nrich information out of the image? We argue that by using visual clues to\nbridge large pretrained vision foundation models and language models, we can do\nso without any extra cross-modal training. Thanks to the strong zero-shot\ncapability of foundation models, we start by constructing a rich semantic\nrepresentation of the image (e.g., image tags, object attributes / locations,\ncaptions) as a structured textual prompt, called visual clues, using a vision\nfoundation model. Based on visual clues, we use large language model to produce\na series of comprehensive descriptions for the visual content, which is then\nverified by the vision model again to select the candidate that aligns best\nwith the image. We evaluate the quality of generated descriptions by\nquantitative and qualitative measurement. The results demonstrate the\neffectiveness of such a structured semantic representation.", + "authors": "Yujia Xie, Luowei Zhou, Xiyang Dai, Lu Yuan, Nguyen Bach, Ce Liu, Michael Zeng", + "published": "2022-06-03", + "updated": "2022-09-14", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.AI", + "cs.CL" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2205.12689v2", + "title": "Large Language Models are Few-Shot Clinical Information Extractors", + "abstract": "A long-running goal of the clinical NLP community is the extraction of\nimportant variables trapped in clinical notes. However, roadblocks have\nincluded dataset shift from the general domain and a lack of public clinical\ncorpora and annotations. In this work, we show that large language models, such\nas InstructGPT, perform well at zero- and few-shot information extraction from\nclinical text despite not being trained specifically for the clinical domain.\nWhereas text classification and generation performance have already been\nstudied extensively in such models, here we additionally demonstrate how to\nleverage them to tackle a diverse set of NLP tasks which require more\nstructured outputs, including span identification, token-level sequence\nclassification, and relation extraction. Further, due to the dearth of\navailable data to evaluate these systems, we introduce new datasets for\nbenchmarking few-shot clinical information extraction based on a manual\nre-annotation of the CASI dataset for new tasks. On the clinical extraction\ntasks we studied, the GPT-3 systems significantly outperform existing zero- and\nfew-shot baselines.", + "authors": "Monica Agrawal, Stefan Hegselmann, Hunter Lang, Yoon Kim, David Sontag", + "published": "2022-05-25", + "updated": "2022-11-30", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL", + "cs.AI" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2304.08485v2", + "title": "Visual Instruction Tuning", + "abstract": "Instruction tuning large language models (LLMs) using machine-generated\ninstruction-following data has improved zero-shot capabilities on new tasks,\nbut the idea is less explored in the multimodal field. In this paper, we\npresent the first attempt to use language-only GPT-4 to generate multimodal\nlanguage-image instruction-following data. By instruction tuning on such\ngenerated data, we introduce LLaVA: Large Language and Vision Assistant, an\nend-to-end trained large multimodal model that connects a vision encoder and\nLLM for general-purpose visual and language understanding.Our early experiments\nshow that LLaVA demonstrates impressive multimodel chat abilities, sometimes\nexhibiting the behaviors of multimodal GPT-4 on unseen images/instructions, and\nyields a 85.1% relative score compared with GPT-4 on a synthetic multimodal\ninstruction-following dataset. When fine-tuned on Science QA, the synergy of\nLLaVA and GPT-4 achieves a new state-of-the-art accuracy of 92.53%. We make\nGPT-4 generated visual instruction tuning data, our model and code base\npublicly available.", + "authors": "Haotian Liu, Chunyuan Li, Qingyang Wu, Yong Jae Lee", + "published": "2023-04-17", + "updated": "2023-12-11", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.AI", + "cs.CL", + "cs.LG" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2109.05014v2", + "title": "An Empirical Study of GPT-3 for Few-Shot Knowledge-Based VQA", + "abstract": "Knowledge-based visual question answering (VQA) involves answering questions\nthat require external knowledge not present in the image. Existing methods\nfirst retrieve knowledge from external resources, then reason over the selected\nknowledge, the input image, and question for answer prediction. However, this\ntwo-step approach could lead to mismatches that potentially limit the VQA\nperformance. For example, the retrieved knowledge might be noisy and irrelevant\nto the question, and the re-embedded knowledge features during reasoning might\ndeviate from their original meanings in the knowledge base (KB). To address\nthis challenge, we propose PICa, a simple yet effective method that Prompts\nGPT3 via the use of Image Captions, for knowledge-based VQA. Inspired by\nGPT-3's power in knowledge retrieval and question answering, instead of using\nstructured KBs as in previous work, we treat GPT-3 as an implicit and\nunstructured KB that can jointly acquire and process relevant knowledge.\nSpecifically, we first convert the image into captions (or tags) that GPT-3 can\nunderstand, then adapt GPT-3 to solve the VQA task in a few-shot manner by just\nproviding a few in-context VQA examples. We further boost performance by\ncarefully investigating: (i) what text formats best describe the image content,\nand (ii) how in-context examples can be better selected and used. PICa unlocks\nthe first use of GPT-3 for multimodal tasks. By using only 16 examples, PICa\nsurpasses the supervised state of the art by an absolute +8.6 points on the\nOK-VQA dataset. We also benchmark PICa on VQAv2, where PICa also shows a decent\nfew-shot performance.", + "authors": "Zhengyuan Yang, Zhe Gan, Jianfeng Wang, Xiaowei Hu, Yumao Lu, Zicheng Liu, Lijuan Wang", + "published": "2021-09-10", + "updated": "2022-09-14", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2308.04005v1", + "title": "Few-shot medical image classification with simple shape and texture text descriptors using vision-language models", + "abstract": "In this work, we investigate the usefulness of vision-language models (VLMs)\nand large language models for binary few-shot classification of medical images.\nWe utilize the GPT-4 model to generate text descriptors that encapsulate the\nshape and texture characteristics of objects in medical images. Subsequently,\nthese GPT-4 generated descriptors, alongside VLMs pre-trained on natural\nimages, are employed to classify chest X-rays and breast ultrasound images. Our\nresults indicate that few-shot classification of medical images using VLMs and\nGPT-4 generated descriptors is a viable approach. However, accurate\nclassification requires to exclude certain descriptors from the calculations of\nthe classification scores. Moreover, we assess the ability of VLMs to evaluate\nshape features in breast mass ultrasound images. We further investigate the\ndegree of variability among the sets of text descriptors produced by GPT-4. Our\nwork provides several important insights about the application of VLMs for\nmedical image analysis.", + "authors": "Michal Byra, Muhammad Febrian Rachmadi, Henrik Skibbe", + "published": "2023-08-08", + "updated": "2023-08-08", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2307.14334v1", + "title": "Towards Generalist Biomedical AI", + "abstract": "Medicine is inherently multimodal, with rich data modalities spanning text,\nimaging, genomics, and more. Generalist biomedical artificial intelligence (AI)\nsystems that flexibly encode, integrate, and interpret this data at scale can\npotentially enable impactful applications ranging from scientific discovery to\ncare delivery. To enable the development of these models, we first curate\nMultiMedBench, a new multimodal biomedical benchmark. MultiMedBench encompasses\n14 diverse tasks such as medical question answering, mammography and\ndermatology image interpretation, radiology report generation and\nsummarization, and genomic variant calling. We then introduce Med-PaLM\nMultimodal (Med-PaLM M), our proof of concept for a generalist biomedical AI\nsystem. Med-PaLM M is a large multimodal generative model that flexibly encodes\nand interprets biomedical data including clinical language, imaging, and\ngenomics with the same set of model weights. Med-PaLM M reaches performance\ncompetitive with or exceeding the state of the art on all MultiMedBench tasks,\noften surpassing specialist models by a wide margin. We also report examples of\nzero-shot generalization to novel medical concepts and tasks, positive transfer\nlearning across tasks, and emergent zero-shot medical reasoning. To further\nprobe the capabilities and limitations of Med-PaLM M, we conduct a radiologist\nevaluation of model-generated (and human) chest X-ray reports and observe\nencouraging performance across model scales. In a side-by-side ranking on 246\nretrospective chest X-rays, clinicians express a pairwise preference for\nMed-PaLM M reports over those produced by radiologists in up to 40.50% of\ncases, suggesting potential clinical utility. While considerable work is needed\nto validate these models in real-world use cases, our results represent a\nmilestone towards the development of generalist biomedical AI systems.", + "authors": "Tao Tu, Shekoofeh Azizi, Danny Driess, Mike Schaekermann, Mohamed Amin, Pi-Chuan Chang, Andrew Carroll, Chuck Lau, Ryutaro Tanno, Ira Ktena, Basil Mustafa, Aakanksha Chowdhery, Yun Liu, Simon Kornblith, David Fleet, Philip Mansfield, Sushant Prakash, Renee Wong, Sunny Virmani, Christopher Semturs, S Sara Mahdavi, Bradley Green, Ewa Dominowska, Blaise Aguera y Arcas, Joelle Barral, Dale Webster, Greg S. Corrado, Yossi Matias, Karan Singhal, Pete Florence, Alan Karthikesalingam, Vivek Natarajan", + "published": "2023-07-26", + "updated": "2023-07-26", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL", + "cs.CV" + ], + "label": "Related Work" + }, + { + "url": "http://arxiv.org/abs/2211.09699v4", + "title": "PromptCap: Prompt-Guided Task-Aware Image Captioning", + "abstract": "Knowledge-based visual question answering (VQA) involves questions that\nrequire world knowledge beyond the image to yield the correct answer. Large\nlanguage models (LMs) like GPT-3 are particularly helpful for this task because\nof their strong knowledge retrieval and reasoning capabilities. To enable LM to\nunderstand images, prior work uses a captioning model to convert images into\ntext. However, when summarizing an image in a single caption sentence, which\nvisual entities to describe are often underspecified. Generic image captions\noften miss visual details essential for the LM to answer visual questions\ncorrectly. To address this challenge, we propose PromptCap (Prompt-guided image\nCaptioning), a captioning model designed to serve as a better connector between\nimages and black-box LMs. Different from generic captions, PromptCap takes a\nnatural-language prompt to control the visual entities to describe in the\ngenerated caption. The prompt contains a question that the caption should aid\nin answering. To avoid extra annotation, PromptCap is trained by examples\nsynthesized with GPT-3 and existing datasets. We demonstrate PromptCap's\neffectiveness on an existing pipeline in which GPT-3 is prompted with image\ncaptions to carry out VQA. PromptCap outperforms generic captions by a large\nmargin and achieves state-of-the-art accuracy on knowledge-based VQA tasks\n(60.4% on OK-VQA and 59.6% on A-OKVQA). Zero-shot results on WebQA show that\nPromptCap generalizes well to unseen domains.", + "authors": "Yushi Hu, Hang Hua, Zhengyuan Yang, Weijia Shi, Noah A Smith, Jiebo Luo", + "published": "2022-11-15", + "updated": "2023-08-17", + "primary_cat": "cs.CV", + "cats": [ + "cs.CV", + "cs.CL" + ], + "label": "Related Work" + }, + { + "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/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/2404.06003v1", + "title": "FreeEval: A Modular Framework for Trustworthy and Efficient Evaluation of Large Language Models", + "abstract": "The rapid development of large language model (LLM) evaluation methodologies\nand datasets has led to a profound challenge: integrating state-of-the-art\nevaluation techniques cost-effectively while ensuring reliability,\nreproducibility, and efficiency. Currently, there is a notable absence of a\nunified and adaptable framework that seamlessly integrates various evaluation\napproaches. Moreover, the reliability of evaluation findings is often\nquestionable due to potential data contamination, with the evaluation\nefficiency commonly overlooked when facing the substantial costs associated\nwith LLM inference. In response to these challenges, we introduce FreeEval, a\nmodular and scalable framework crafted to enable trustworthy and efficient\nautomatic evaluations of LLMs. Firstly, FreeEval's unified abstractions\nsimplify the integration and improve the transparency of diverse evaluation\nmethodologies, encompassing dynamic evaluation that demand sophisticated LLM\ninteractions. Secondly, the framework integrates meta-evaluation techniques\nlike human evaluation and data contamination detection, which, along with\ndynamic evaluation modules in the platform, enhance the fairness of the\nevaluation outcomes. Lastly, FreeEval is designed with a high-performance\ninfrastructure, including distributed computation and caching strategies,\nenabling extensive evaluations across multi-node, multi-GPU clusters for\nopen-source and proprietary LLMs.", + "authors": "Zhuohao Yu, Chang Gao, Wenjin Yao, Yidong Wang, Zhengran Zeng, Wei Ye, Jindong Wang, Yue Zhang, Shikun Zhang", + "published": "2024-04-09", + "updated": "2024-04-09", + "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/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/2309.08836v2", + "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/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/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/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/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/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/2310.18333v3", + "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/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/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/2402.11764v1", + "title": "ChatGPT Based Data Augmentation for Improved Parameter-Efficient Debiasing of LLMs", + "abstract": "Large Language models (LLMs), while powerful, exhibit harmful social biases.\nDebiasing is often challenging due to computational costs, data constraints,\nand potential degradation of multi-task language capabilities. This work\nintroduces a novel approach utilizing ChatGPT to generate synthetic training\ndata, aiming to enhance the debiasing of LLMs. We propose two strategies:\nTargeted Prompting, which provides effective debiasing for known biases but\nnecessitates prior specification of bias in question; and General Prompting,\nwhich, while slightly less effective, offers debiasing across various\ncategories. We leverage resource-efficient LLM debiasing using adapter tuning\nand compare the effectiveness of our synthetic data to existing debiasing\ndatasets. Our results reveal that: (1) ChatGPT can efficiently produce\nhigh-quality training data for debiasing other LLMs; (2) data produced via our\napproach surpasses existing datasets in debiasing performance while also\npreserving internal knowledge of a pre-trained LLM; and (3) synthetic data\nexhibits generalizability across categories, effectively mitigating various\nbiases, including intersectional ones. These findings underscore the potential\nof synthetic data in advancing the fairness of LLMs with minimal retraining\ncost.", + "authors": "Pengrui Han, Rafal Kocielnik, Adhithya Saravanan, Roy Jiang, Or Sharir, Anima Anandkumar", + "published": "2024-02-19", + "updated": "2024-02-19", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL", + "cs.AI", + "cs.CY", + "68T50", + "I.2.7; K.4.1" + ], + "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/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/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/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/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/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/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/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/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/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/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/2401.11033v4", + "title": "FAIR Enough: How Can We Develop and Assess a FAIR-Compliant Dataset for Large Language Models' Training?", + "abstract": "The rapid evolution of Large Language Models (LLMs) highlights the necessity\nfor ethical considerations and data integrity in AI development, particularly\nemphasizing the role of FAIR (Findable, Accessible, Interoperable, Reusable)\ndata principles. While these principles are crucial for ethical data\nstewardship, their specific application in the context of LLM training data\nremains an under-explored area. This research gap is the focus of our study,\nwhich begins with an examination of existing literature to underline the\nimportance of FAIR principles in managing data for LLM training. Building upon\nthis, we propose a novel framework designed to integrate FAIR principles into\nthe LLM development lifecycle. A contribution of our work is the development of\na comprehensive checklist intended to guide researchers and developers in\napplying FAIR data principles consistently across the model development\nprocess. The utility and effectiveness of our framework are validated through a\ncase study on creating a FAIR-compliant dataset aimed at detecting and\nmitigating biases in LLMs. We present this framework to the community as a tool\nto foster the creation of technologically advanced, ethically grounded, and\nsocially responsible AI models.", + "authors": "Shaina Raza, Shardul Ghuge, Chen Ding, Elham Dolatabadi, Deval Pandya", + "published": "2024-01-19", + "updated": "2024-04-03", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL" + ], + "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/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/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/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/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/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/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/2305.03514v3", + "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/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/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/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/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/2311.18580v1", + "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/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/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/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/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/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/2311.06899v4", + "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/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/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/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/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/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/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/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/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/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/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/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" + }, + { + "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/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/2307.11761v1", + "title": "Fairness of ChatGPT and the Role Of Explainable-Guided Prompts", + "abstract": "Our research investigates the potential of Large-scale Language Models\n(LLMs), specifically OpenAI's GPT, in credit risk assessment-a binary\nclassification task. Our findings suggest that LLMs, when directed by\njudiciously designed prompts and supplemented with domain-specific knowledge,\ncan parallel the performance of traditional Machine Learning (ML) models.\nIntriguingly, they achieve this with significantly less data-40 times less,\nutilizing merely 20 data points compared to the ML's 800. LLMs particularly\nexcel in minimizing false positives and enhancing fairness, both being vital\naspects of risk analysis. While our results did not surpass those of classical\nML models, they underscore the potential of LLMs in analogous tasks, laying a\ngroundwork for future explorations into harnessing the capabilities of LLMs in\ndiverse ML tasks.", + "authors": "Yashar Deldjoo", + "published": "2023-07-14", + "updated": "2023-07-14", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL", + "cs.AI" + ], + "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/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/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/2402.10567v3", + "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/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/2311.03033v1", + "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/2404.02650v1", + "title": "Towards detecting unanticipated bias in Large Language Models", + "abstract": "Over the last year, Large Language Models (LLMs) like ChatGPT have become\nwidely available and have exhibited fairness issues similar to those in\nprevious machine learning systems. Current research is primarily focused on\nanalyzing and quantifying these biases in training data and their impact on the\ndecisions of these models, alongside developing mitigation strategies. This\nresearch largely targets well-known biases related to gender, race, ethnicity,\nand language. However, it is clear that LLMs are also affected by other, less\nobvious implicit biases. The complex and often opaque nature of these models\nmakes detecting such biases challenging, yet this is crucial due to their\npotential negative impact in various applications. In this paper, we explore\nnew avenues for detecting these unanticipated biases in LLMs, focusing\nspecifically on Uncertainty Quantification and Explainable AI methods. These\napproaches aim to assess the certainty of model decisions and to make the\ninternal decision-making processes of LLMs more transparent, thereby\nidentifying and understanding biases that are not immediately apparent. Through\nthis research, we aim to contribute to the development of fairer and more\ntransparent AI systems.", + "authors": "Anna Kruspe", + "published": "2024-04-03", + "updated": "2024-04-03", + "primary_cat": "cs.LG", + "cats": [ + "cs.LG", + "cs.AI", + "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/2311.18140v1", + "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/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/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/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/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/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/2310.06500v1", + "title": "MetaAgents: Simulating Interactions of Human Behaviors for LLM-based Task-oriented Coordination via Collaborative Generative Agents", + "abstract": "Significant advancements have occurred in the application of Large Language\nModels (LLMs) for various tasks and social simulations. Despite this, their\ncapacities to coordinate within task-oriented social contexts are\nunder-explored. Such capabilities are crucial if LLMs are to effectively mimic\nhuman-like social behavior and produce meaningful results. To bridge this gap,\nwe introduce collaborative generative agents, endowing LLM-based Agents with\nconsistent behavior patterns and task-solving abilities. We situate these\nagents in a simulated job fair environment as a case study to scrutinize their\ncoordination skills. We propose a novel framework that equips collaborative\ngenerative agents with human-like reasoning abilities and specialized skills.\nOur evaluation demonstrates that these agents show promising performance.\nHowever, we also uncover limitations that hinder their effectiveness in more\ncomplex coordination tasks. Our work provides valuable insights into the role\nand evolution of LLMs in task-oriented social simulations.", + "authors": "Yuan Li, Yixuan Zhang, Lichao Sun", + "published": "2023-10-10", + "updated": "2023-10-10", + "primary_cat": "cs.AI", + "cats": [ + "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/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/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/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/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/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/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/2310.13343v1", + "title": "Challenges and Contributing Factors in the Utilization of Large Language Models (LLMs)", + "abstract": "With the development of large language models (LLMs) like the GPT series,\ntheir widespread use across various application scenarios presents a myriad of\nchallenges. This review initially explores the issue of domain specificity,\nwhere LLMs may struggle to provide precise answers to specialized questions\nwithin niche fields. The problem of knowledge forgetting arises as these LLMs\nmight find it hard to balance old and new information. The knowledge repetition\nphenomenon reveals that sometimes LLMs might deliver overly mechanized\nresponses, lacking depth and originality. Furthermore, knowledge illusion\ndescribes situations where LLMs might provide answers that seem insightful but\nare actually superficial, while knowledge toxicity focuses on harmful or biased\ninformation outputs. These challenges underscore problems in the training data\nand algorithmic design of LLMs. To address these issues, it's suggested to\ndiversify training data, fine-tune models, enhance transparency and\ninterpretability, and incorporate ethics and fairness training. Future\ntechnological trends might lean towards iterative methodologies, multimodal\nlearning, model personalization and customization, and real-time learning and\nfeedback mechanisms. In conclusion, future LLMs should prioritize fairness,\ntransparency, and ethics, ensuring they uphold high moral and ethical standards\nwhen serving humanity.", + "authors": "Xiaoliang Chen, Liangbin Li, Le Chang, Yunhe Huang, Yuxuan Zhao, Yuxiao Zhang, Dinuo Li", + "published": "2023-10-20", + "updated": "2023-10-20", + "primary_cat": "cs.CL", + "cats": [ + "cs.CL", + "cs.AI" + ], + "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/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" + } +] \ No newline at end of file