nielsr HF Staff commited on
Commit
32af4ff
·
verified ·
1 Parent(s): c80f1c3

Improve model card metadata and add project links

Browse files

This PR improves the model card for better documentation of the model:

* Sets the correct pipeline tag (`text-ranking`) for the model.
* Specifies the library used (`sentence-transformers`).
* Adds links to the project page, paper, and code repository for quick access.

Files changed (1) hide show
  1. README.md +52 -341
README.md CHANGED
@@ -1,4 +1,12 @@
1
  ---
 
 
 
 
 
 
 
 
2
  tags:
3
  - sentence-transformers
4
  - sentence-similarity
@@ -6,76 +14,65 @@ tags:
6
  - generated_from_trainer
7
  - dataset_size:784827
8
  - loss:ContrastiveLoss
9
- base_model: intfloat/e5-large-v2
10
  widget:
11
- - source_sentence: >-
12
- query: The study addresses the need for effective tools that allow both
13
- novice and expert users to analyze the diversity of news coverage about
14
- events. It highlights the importance of tailoring the interface to
15
- accommodate non-expert users while also considering the insights of
16
- journalism-savvy users, indicating a gap in existing systems that cater to
17
- varying levels of expertise in news analysis.We suggest combining 'a
18
- coordinated visualization interface tailored for visualization non-expert
19
- users' and
20
  sentences:
21
  - graph convolution
22
  - Monte-Carlo sampling
23
  - geometric features derived from perception sensor data
24
- - source_sentence: >-
25
- query: The accuracy of pixel flows is crucial for achieving high-quality
26
- video enhancement, yet most prior works focus on estimating dense flows that
27
- are generally less robust and computationally expensive. This highlights a
28
- gap in existing methodologies that fail to prioritize accuracy over density,
29
- necessitating a more efficient approach to flow estimation for video
30
- enhancement tasks.We suggest combining 'sparse point cloud data' and
31
  sentences:
32
  - a human cognition mechanism, object unity
33
  - Bayesian Optimization
34
  - offline supervised learning
35
- - source_sentence: >-
36
- query: The traditional frame of discernment lacks a crucial factor, the
37
- sequence of propositions, which limits the effectiveness of existing methods
38
- to measure uncertainty. This gap highlights the need for a more
39
- comprehensive approach that can better represent the relationships between
40
- the elements of the frame of discernment.We suggest 'combine the order of
41
- propositions and the mass of them' inspired by
42
  sentences:
43
  - a MIA-Module
44
  - an Explore-m problem--a well-studied problem related to multi-armed bandits
45
  - based on the novel method UGPIG
46
- - source_sentence: >-
47
- query: Existing methods for anomaly detection on dynamic graphs struggle
48
- with capturing complex time information in graph structures and generating
49
- effective negative samples for unsupervised learning. These challenges
50
- highlight the need for improved methodologies that can address the
51
- limitations of current approaches in this field.We suggest combining 'a
52
- message-passing framework' and
53
  sentences:
54
  - an LSTM encoder-decoder
55
  - an energy-based model
56
- - >-
57
- learning the frame-wise associations between detections in consecutive
58
- frames
59
- - source_sentence: >-
60
- query: The study addresses the need for effective time series forecasting
61
- methods to estimate the spread of epidemics, particularly in light of the
62
- resurgence of COVID-19 cases. It highlights the importance of accurately
63
- modeling both linear and non-linear features of epidemic data to provide
64
- state authorities and health officials with reliable short-term forecasts
65
- and strategies.We suggest combining 'ARIMA' and
66
  sentences:
67
  - visualization methodologies
68
  - geometry
69
  - the utilization of a gradient signed distance field (gradient-SDF)
70
- pipeline_tag: sentence-similarity
71
- library_name: sentence-transformers
72
- license: cc
73
- datasets:
74
- - noystl/Recombination-Pred
75
- language:
76
- - en
77
  ---
78
 
 
 
 
 
 
79
  # SentenceTransformer based on intfloat/e5-large-v2
80
 
81
  This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/e5-large-v2](https://huggingface.co/intfloat/e5-large-v2). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
@@ -429,8 +426,8 @@ You can finetune this model on your own dataset.
429
  | 0.7013 | 8600 | 0.0033 |
430
  | 0.7095 | 8700 | 0.0031 |
431
  | 0.7176 | 8800 | 0.0029 |
432
- | 0.7258 | 8900 | 0.0037 |
433
- | 0.7339 | 9000 | 0.0034 |
434
  | 0.7421 | 9100 | 0.0031 |
435
  | 0.7502 | 9200 | 0.003 |
436
  | 0.7584 | 9300 | 0.0031 |
@@ -483,304 +480,18 @@ You can finetune this model on your own dataset.
483
  | 1.1416 | 14000 | 0.0034 |
484
  | 1.1498 | 14100 | 0.0031 |
485
  | 1.1580 | 14200 | 0.0029 |
486
- | 1.1661 | 14300 | 0.0029 |
487
  | 1.1743 | 14400 | 0.0028 |
488
  | 1.1824 | 14500 | 0.0037 |
489
  | 1.1906 | 14600 | 0.0029 |
490
- | 1.1987 | 14700 | 0.0028 |
491
  | 1.2069 | 14800 | 0.0029 |
492
  | 1.2150 | 14900 | 0.0035 |
493
  | 1.2232 | 15000 | 0.0029 |
494
- | 1.2313 | 15100 | 0.0029 |
495
  | 1.2395 | 15200 | 0.0027 |
496
  | 1.2477 | 15300 | 0.003 |
497
- | 1.2558 | 15400 | 0.0035 |
498
  | 1.2640 | 15500 | 0.0027 |
499
  | 1.2721 | 15600 | 0.0028 |
500
- | 1.2803 | 15700 | 0.0028 |
501
- | 1.2884 | 15800 | 0.0037 |
502
- | 1.2966 | 15900 | 0.0028 |
503
- | 1.3047 | 16000 | 0.0028 |
504
- | 1.3129 | 16100 | 0.0028 |
505
- | 1.3210 | 16200 | 0.0029 |
506
- | 1.3292 | 16300 | 0.0034 |
507
- | 1.3374 | 16400 | 0.0028 |
508
- | 1.3455 | 16500 | 0.0026 |
509
- | 1.3537 | 16600 | 0.0029 |
510
- | 1.3618 | 16700 | 0.0034 |
511
- | 1.3700 | 16800 | 0.0028 |
512
- | 1.3781 | 16900 | 0.0027 |
513
- | 1.3863 | 17000 | 0.003 |
514
- | 1.3944 | 17100 | 0.0034 |
515
- | 1.4026 | 17200 | 0.0028 |
516
- | 1.4107 | 17300 | 0.0028 |
517
- | 1.4189 | 17400 | 0.0027 |
518
- | 1.4271 | 17500 | 0.0028 |
519
- | 1.4352 | 17600 | 0.0036 |
520
- | 1.4434 | 17700 | 0.0028 |
521
- | 1.4515 | 17800 | 0.0027 |
522
- | 1.4597 | 17900 | 0.0028 |
523
- | 1.4678 | 18000 | 0.0032 |
524
- | 1.4760 | 18100 | 0.0029 |
525
- | 1.4841 | 18200 | 0.0028 |
526
- | 1.4923 | 18300 | 0.0028 |
527
- | 1.5004 | 18400 | 0.0028 |
528
- | 1.5086 | 18500 | 0.0033 |
529
- | 1.5168 | 18600 | 0.0026 |
530
- | 1.5249 | 18700 | 0.0027 |
531
- | 1.5331 | 18800 | 0.0028 |
532
- | 1.5412 | 18900 | 0.0035 |
533
- | 1.5494 | 19000 | 0.0026 |
534
- | 1.5575 | 19100 | 0.0027 |
535
- | 1.5657 | 19200 | 0.0027 |
536
- | 1.5738 | 19300 | 0.0028 |
537
- | 1.5820 | 19400 | 0.0033 |
538
- | 1.5901 | 19500 | 0.0026 |
539
- | 1.5983 | 19600 | 0.0028 |
540
- | 1.6065 | 19700 | 0.0026 |
541
- | 1.6146 | 19800 | 0.0033 |
542
- | 1.6228 | 19900 | 0.0026 |
543
- | 1.6309 | 20000 | 0.0027 |
544
- | 1.6391 | 20100 | 0.0029 |
545
- | 1.6472 | 20200 | 0.0032 |
546
- | 1.6554 | 20300 | 0.0028 |
547
- | 1.6635 | 20400 | 0.0025 |
548
- | 1.6717 | 20500 | 0.0025 |
549
- | 1.6798 | 20600 | 0.0025 |
550
- | 1.6880 | 20700 | 0.003 |
551
- | 1.6962 | 20800 | 0.0028 |
552
- | 1.7043 | 20900 | 0.0026 |
553
- | 1.7125 | 21000 | 0.0024 |
554
- | 1.7206 | 21100 | 0.0028 |
555
- | 1.7288 | 21200 | 0.0028 |
556
- | 1.7369 | 21300 | 0.0026 |
557
- | 1.7451 | 21400 | 0.0026 |
558
- | 1.7532 | 21500 | 0.0025 |
559
- | 1.7614 | 21600 | 0.003 |
560
- | 1.7696 | 21700 | 0.0027 |
561
- | 1.7777 | 21800 | 0.0023 |
562
- | 1.7859 | 21900 | 0.0025 |
563
- | 1.7940 | 22000 | 0.0028 |
564
- | 1.8022 | 22100 | 0.0025 |
565
- | 1.8103 | 22200 | 0.0026 |
566
- | 1.8185 | 22300 | 0.0024 |
567
- | 1.8266 | 22400 | 0.0025 |
568
- | 1.8348 | 22500 | 0.0029 |
569
- | 1.8429 | 22600 | 0.0028 |
570
- | 1.8511 | 22700 | 0.0024 |
571
- | 1.8593 | 22800 | 0.0026 |
572
- | 1.8674 | 22900 | 0.003 |
573
- | 1.8756 | 23000 | 0.0026 |
574
- | 1.8837 | 23100 | 0.0025 |
575
- | 1.8919 | 23200 | 0.0025 |
576
- | 1.9000 | 23300 | 0.0027 |
577
- | 1.9082 | 23400 | 0.0025 |
578
- | 1.9163 | 23500 | 0.0026 |
579
- | 1.9245 | 23600 | 0.0026 |
580
- | 1.9326 | 23700 | 0.0026 |
581
- | 1.9408 | 23800 | 0.003 |
582
- | 1.9490 | 23900 | 0.0026 |
583
- | 1.9571 | 24000 | 0.0026 |
584
- | 1.9653 | 24100 | 0.0025 |
585
- | 1.9734 | 24200 | 0.003 |
586
- | 1.9816 | 24300 | 0.0028 |
587
- | 1.9897 | 24400 | 0.0025 |
588
- | 1.9979 | 24500 | 0.0028 |
589
- | 2.0060 | 24600 | 0.0029 |
590
- | 2.0142 | 24700 | 0.0025 |
591
- | 2.0223 | 24800 | 0.0026 |
592
- | 2.0305 | 24900 | 0.0031 |
593
- | 2.0387 | 25000 | 0.0025 |
594
- | 2.0468 | 25100 | 0.0025 |
595
- | 2.0550 | 25200 | 0.0023 |
596
- | 2.0631 | 25300 | 0.0024 |
597
- | 2.0713 | 25400 | 0.0031 |
598
- | 2.0794 | 25500 | 0.0024 |
599
- | 2.0876 | 25600 | 0.0025 |
600
- | 2.0957 | 25700 | 0.0024 |
601
- | 2.1039 | 25800 | 0.0031 |
602
- | 2.1120 | 25900 | 0.0024 |
603
- | 2.1202 | 26000 | 0.0025 |
604
- | 2.1284 | 26100 | 0.0025 |
605
- | 2.1365 | 26200 | 0.0024 |
606
- | 2.1447 | 26300 | 0.003 |
607
- | 2.1528 | 26400 | 0.0025 |
608
- | 2.1610 | 26500 | 0.0024 |
609
- | 2.1691 | 26600 | 0.0026 |
610
- | 2.1773 | 26700 | 0.003 |
611
- | 2.1854 | 26800 | 0.0025 |
612
- | 2.1936 | 26900 | 0.0025 |
613
- | 2.2017 | 27000 | 0.0024 |
614
- | 2.2099 | 27100 | 0.003 |
615
- | 2.2181 | 27200 | 0.0024 |
616
- | 2.2262 | 27300 | 0.0026 |
617
- | 2.2344 | 27400 | 0.0023 |
618
- | 2.2425 | 27500 | 0.0023 |
619
- | 2.2507 | 27600 | 0.0031 |
620
- | 2.2588 | 27700 | 0.0023 |
621
- | 2.2670 | 27800 | 0.0022 |
622
- | 2.2751 | 27900 | 0.0024 |
623
- | 2.2833 | 28000 | 0.0032 |
624
- | 2.2914 | 28100 | 0.0024 |
625
- | 2.2996 | 28200 | 0.0023 |
626
- | 2.3078 | 28300 | 0.0026 |
627
- | 2.3159 | 28400 | 0.0023 |
628
- | 2.3241 | 28500 | 0.0031 |
629
- | 2.3322 | 28600 | 0.0024 |
630
- | 2.3404 | 28700 | 0.0023 |
631
- | 2.3485 | 28800 | 0.0023 |
632
- | 2.3567 | 28900 | 0.0031 |
633
- | 2.3648 | 29000 | 0.0024 |
634
- | 2.3730 | 29100 | 0.0023 |
635
- | 2.3811 | 29200 | 0.0025 |
636
- | 2.3893 | 29300 | 0.0027 |
637
- | 2.3975 | 29400 | 0.0029 |
638
- | 2.4056 | 29500 | 0.0022 |
639
- | 2.4138 | 29600 | 0.0024 |
640
- | 2.4219 | 29700 | 0.0023 |
641
- | 2.4301 | 29800 | 0.0031 |
642
- | 2.4382 | 29900 | 0.0024 |
643
- | 2.4464 | 30000 | 0.0023 |
644
- | 2.4545 | 30100 | 0.0022 |
645
- | 2.4627 | 30200 | 0.0029 |
646
- | 2.4708 | 30300 | 0.0024 |
647
- | 2.4790 | 30400 | 0.0025 |
648
- | 2.4872 | 30500 | 0.0024 |
649
- | 2.4953 | 30600 | 0.0024 |
650
- | 2.5035 | 30700 | 0.003 |
651
- | 2.5116 | 30800 | 0.0021 |
652
- | 2.5198 | 30900 | 0.0023 |
653
- | 2.5279 | 31000 | 0.0024 |
654
- | 2.5361 | 31100 | 0.0032 |
655
- | 2.5442 | 31200 | 0.0023 |
656
- | 2.5524 | 31300 | 0.0022 |
657
- | 2.5605 | 31400 | 0.0024 |
658
- | 2.5687 | 31500 | 0.0023 |
659
- | 2.5769 | 31600 | 0.0029 |
660
- | 2.5850 | 31700 | 0.0023 |
661
- | 2.5932 | 31800 | 0.0023 |
662
- | 2.6013 | 31900 | 0.0023 |
663
- | 2.6095 | 32000 | 0.003 |
664
- | 2.6176 | 32100 | 0.0023 |
665
- | 2.6258 | 32200 | 0.0023 |
666
- | 2.6339 | 32300 | 0.0024 |
667
- | 2.6421 | 32400 | 0.0027 |
668
- | 2.6502 | 32500 | 0.0028 |
669
- | 2.6584 | 32600 | 0.0023 |
670
- | 2.6666 | 32700 | 0.0021 |
671
- | 2.6747 | 32800 | 0.0023 |
672
- | 2.6829 | 32900 | 0.0026 |
673
- | 2.6910 | 33000 | 0.0024 |
674
- | 2.6992 | 33100 | 0.0023 |
675
- | 2.7073 | 33200 | 0.0023 |
676
- | 2.7155 | 33300 | 0.0024 |
677
- | 2.7236 | 33400 | 0.0024 |
678
- | 2.7318 | 33500 | 0.0024 |
679
- | 2.7399 | 33600 | 0.0023 |
680
- | 2.7481 | 33700 | 0.0022 |
681
- | 2.7563 | 33800 | 0.0027 |
682
- | 2.7644 | 33900 | 0.0023 |
683
- | 2.7726 | 34000 | 0.0023 |
684
- | 2.7807 | 34100 | 0.0021 |
685
- | 2.7889 | 34200 | 0.0025 |
686
- | 2.7970 | 34300 | 0.0022 |
687
- | 2.8052 | 34400 | 0.0022 |
688
- | 2.8133 | 34500 | 0.0021 |
689
- | 2.8215 | 34600 | 0.0022 |
690
- | 2.8297 | 34700 | 0.0026 |
691
- | 2.8378 | 34800 | 0.0024 |
692
- | 2.8460 | 34900 | 0.0023 |
693
- | 2.8541 | 35000 | 0.0022 |
694
- | 2.8623 | 35100 | 0.0026 |
695
- | 2.8704 | 35200 | 0.0023 |
696
- | 2.8786 | 35300 | 0.0022 |
697
- | 2.8867 | 35400 | 0.0023 |
698
- | 2.8949 | 35500 | 0.0022 |
699
- | 2.9030 | 35600 | 0.0025 |
700
- | 2.9112 | 35700 | 0.0023 |
701
- | 2.9194 | 35800 | 0.0022 |
702
- | 2.9275 | 35900 | 0.0022 |
703
- | 2.9357 | 36000 | 0.0028 |
704
- | 2.9438 | 36100 | 0.0022 |
705
- | 2.9520 | 36200 | 0.0023 |
706
- | 2.9601 | 36300 | 0.0022 |
707
- | 2.9683 | 36400 | 0.0026 |
708
- | 2.9764 | 36500 | 0.0024 |
709
- | 2.9846 | 36600 | 0.0024 |
710
- | 2.9927 | 36700 | 0.0023 |
711
-
712
- </details>
713
-
714
- ### Framework Versions
715
- - Python: 3.11.2
716
- - Sentence Transformers: 3.3.1
717
- - Transformers: 4.49.0
718
- - PyTorch: 2.5.1+cu124
719
- - Accelerate: 1.0.1
720
- - Datasets: 3.1.0
721
- - Tokenizers: 0.21.0
722
-
723
- ## Citation
724
-
725
- ### BibTeX
726
- ```bibtex
727
- @misc{sternlicht2025chimeraknowledgebaseidea,
728
- title={CHIMERA: A Knowledge Base of Idea Recombination in Scientific Literature},
729
- author={Noy Sternlicht and Tom Hope},
730
- year={2025},
731
- eprint={2505.20779},
732
- archivePrefix={arXiv},
733
- primaryClass={cs.CL},
734
- url={https://arxiv.org/abs/2505.20779},
735
- }
736
- ```
737
-
738
- #### Sentence Transformers
739
- ```bibtex
740
- @inproceedings{reimers-2019-sentence-bert,
741
- title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
742
- author = "Reimers, Nils and Gurevych, Iryna",
743
- booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
744
- month = "11",
745
- year = "2019",
746
- publisher = "Association for Computational Linguistics",
747
- url = "https://arxiv.org/abs/1908.10084",
748
- }
749
- ```
750
-
751
- #### ContrastiveLoss
752
- ```bibtex
753
- @inproceedings{hadsell2006dimensionality,
754
- author={Hadsell, R. and Chopra, S. and LeCun, Y.},
755
- booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
756
- title={Dimensionality Reduction by Learning an Invariant Mapping},
757
- year={2006},
758
- volume={2},
759
- number={},
760
- pages={1735-1742},
761
- doi={10.1109/CVPR.2006.100}
762
- }
763
- ```
764
-
765
- **Quick Links**
766
- - 🌐 [Project](https://noy-sternlicht.github.io/CHIMERA-Web)
767
- - 📃 [Paper](https://arxiv.org/abs/2505.20779)
768
- - 🛠️ [Code](https://github.com/noy-sternlicht/CHIMERA-KB)
769
-
770
- <!--
771
- ## Glossary
772
-
773
- *Clearly define terms in order to be accessible across audiences.*
774
- -->
775
-
776
- <!--
777
- ## Model Card Authors
778
-
779
- *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
780
- -->
781
-
782
- <!--
783
- ## Model Card Contact
784
-
785
- *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
786
- -->
 
1
  ---
2
+ base_model: intfloat/e5-large-v2
3
+ datasets:
4
+ - noystl/Recombination-Pred
5
+ language:
6
+ - en
7
+ library_name: sentence-transformers
8
+ license: cc
9
+ pipeline_tag: text-ranking
10
  tags:
11
  - sentence-transformers
12
  - sentence-similarity
 
14
  - generated_from_trainer
15
  - dataset_size:784827
16
  - loss:ContrastiveLoss
 
17
  widget:
18
+ - source_sentence: 'query: The study addresses the need for effective tools that allow
19
+ both novice and expert users to analyze the diversity of news coverage about events.
20
+ It highlights the importance of tailoring the interface to accommodate non-expert
21
+ users while also considering the insights of journalism-savvy users, indicating
22
+ a gap in existing systems that cater to varying levels of expertise in news analysis.We
23
+ suggest combining ''a coordinated visualization interface tailored for visualization
24
+ non-expert users'' and '
 
 
25
  sentences:
26
  - graph convolution
27
  - Monte-Carlo sampling
28
  - geometric features derived from perception sensor data
29
+ - source_sentence: 'query: The accuracy of pixel flows is crucial for achieving high-quality
30
+ video enhancement, yet most prior works focus on estimating dense flows that are
31
+ generally less robust and computationally expensive. This highlights a gap in
32
+ existing methodologies that fail to prioritize accuracy over density, necessitating
33
+ a more efficient approach to flow estimation for video enhancement tasks.We suggest
34
+ combining ''sparse point cloud data'' and '
 
35
  sentences:
36
  - a human cognition mechanism, object unity
37
  - Bayesian Optimization
38
  - offline supervised learning
39
+ - source_sentence: 'query: The traditional frame of discernment lacks a crucial factor,
40
+ the sequence of propositions, which limits the effectiveness of existing methods
41
+ to measure uncertainty. This gap highlights the need for a more comprehensive
42
+ approach that can better represent the relationships between the elements of the
43
+ frame of discernment.We suggest ''combine the order of propositions and the mass
44
+ of them'' inspired by '
 
45
  sentences:
46
  - a MIA-Module
47
  - an Explore-m problem--a well-studied problem related to multi-armed bandits
48
  - based on the novel method UGPIG
49
+ - source_sentence: 'query: Existing methods for anomaly detection on dynamic graphs
50
+ struggle with capturing complex time information in graph structures and generating
51
+ effective negative samples for unsupervised learning. These challenges highlight
52
+ the need for improved methodologies that can address the limitations of current
53
+ approaches in this field.We suggest combining ''a message-passing framework''
54
+ and '
 
55
  sentences:
56
  - an LSTM encoder-decoder
57
  - an energy-based model
58
+ - learning the frame-wise associations between detections in consecutive frames
59
+ - source_sentence: 'query: The study addresses the need for effective time series
60
+ forecasting methods to estimate the spread of epidemics, particularly in light
61
+ of the resurgence of COVID-19 cases. It highlights the importance of accurately
62
+ modeling both linear and non-linear features of epidemic data to provide state
63
+ authorities and health officials with reliable short-term forecasts and strategies.We
64
+ suggest combining ''ARIMA'' and '
 
 
 
65
  sentences:
66
  - visualization methodologies
67
  - geometry
68
  - the utilization of a gradient signed distance field (gradient-SDF)
 
 
 
 
 
 
 
69
  ---
70
 
71
+ **Quick Links**
72
+ - 🌐 [Project](https://noy-sternlicht.github.io/CHIMERA-Web)
73
+ - 📃 [Paper](https://arxiv.org/abs/2505.20779)
74
+ - 🛠️ [Code](https://github.com/noy-sternlicht/CHIMERA-KB)
75
+
76
  # SentenceTransformer based on intfloat/e5-large-v2
77
 
78
  This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/e5-large-v2](https://huggingface.co/intfloat/e5-large-v2). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
 
426
  | 0.7013 | 8600 | 0.0033 |
427
  | 0.7095 | 8700 | 0.0031 |
428
  | 0.7176 | 8800 | 0.0029 |
429
+ | 0.7258 | 8900 | 0.0036 |
430
+ | 0.7339 | 9000 | 0.0033 |
431
  | 0.7421 | 9100 | 0.0031 |
432
  | 0.7502 | 9200 | 0.003 |
433
  | 0.7584 | 9300 | 0.0031 |
 
480
  | 1.1416 | 14000 | 0.0034 |
481
  | 1.1498 | 14100 | 0.0031 |
482
  | 1.1580 | 14200 | 0.0029 |
483
+ | 1.1661 | 14300 | 0.0027 |
484
  | 1.1743 | 14400 | 0.0028 |
485
  | 1.1824 | 14500 | 0.0037 |
486
  | 1.1906 | 14600 | 0.0029 |
487
+ | 1.1987 | 14700 | 0.0027 |
488
  | 1.2069 | 14800 | 0.0029 |
489
  | 1.2150 | 14900 | 0.0035 |
490
  | 1.2232 | 15000 | 0.0029 |
491
+ | 1.2313 | 15100 | 0.0028 |
492
  | 1.2395 | 15200 | 0.0027 |
493
  | 1.2477 | 15300 | 0.003 |
494
+ | 1.2558 | 15400 | 0.0034 |
495
  | 1.2640 | 15500 | 0.0027 |
496
  | 1.2721 | 15600 | 0.0028 |
497
+ | 1.2803 | 15700 | 0.0028 |