{"url":"/task/natural-language-inference","name":"Natural Language Inference","slug":"natural-language-inference","description_markdown":"**Natural language inference (NLI)** is the task of determining whether a \"hypothesis\" is \r\ntrue (entailment), false (contradiction), or undetermined (neutral) given a \"premise\".\r\n\r\nExample:\r\n\r\n| Premise | Label | Hypothesis |\r\n| --- | ---| --- |\r\n| A man inspects the uniform of a figure in some East Asian country. | contradiction | The man is sleeping. |\r\n| An older and younger man smiling. | neutral  | Two men are smiling and laughing at the cats playing on the floor. |\r\n| A soccer game with multiple males playing. | entailment | Some men are playing a sport. |\r\n\r\nApproaches used for NLI include earlier symbolic and statistical approaches to more recent deep learning approaches. Benchmark datasets used for NLI include [SNLI](/dataset/snli), [MultiNLI](/dataset/multinli), [SciTail](/dataset/scitail), among others. You can get hands-on practice on the SNLI task by following this [d2l.ai chapter](https://d2l.ai/chapter_natural-language-processing-applications/natural-language-inference-and-dataset.html).\r\n\r\nFurther readings:\r\n\r\n- [Recent Advances in Natural Language Inference: A Survey of Benchmarks, Resources, and Approaches](https://arxiv.org/abs/1904.01172)","categories":[{"name":"Natural Language Processing","url":"/area/natural-language-processing"},{"name":"Reasoning","url":"/area/reasoning"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":1961,"papers_with_code":821,"benchmarks":33,"benchmark_tables_in_archive":37,"benchmark_tables_shown":37,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":81,"subtasks":3,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/natural-language-inference-on-snli","slug":"natural-language-inference-on-snli","dataset":"SNLI","dataset_url":"/dataset/snli","rows_in_archive":98,"metrics":["% Test Accuracy","% Train Accuracy","Parameters","Dev Accuracy","% Dev Accuracy","Accuracy"],"first_row_in_archive_order":{"model":"UnitedSynT5 (3B)","paper_title":"First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI","paper_url":"/paper/first-train-to-generate-then-generate-to","paper_date":"2024-12-12","arxiv_id":"2412.09263","code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-rte","slug":"natural-language-inference-on-rte","dataset":"RTE","dataset_url":"/dataset/rte","rows_in_archive":90,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Vega v2 6B (KD-based prompt transfer)","paper_title":"Toward Efficient Language Model Pretraining and Downstream Adaptation via Self-Evolution: A Case Study on SuperGLUE","paper_url":"/paper/toward-efficient-language-model-pretraining","paper_date":"2022-12-04","arxiv_id":"2212.01853","code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-multinli","slug":"natural-language-inference-on-multinli","dataset":"MultiNLI","dataset_url":"/dataset/multinli","rows_in_archive":67,"metrics":["Matched","Mismatched","Accuracy","Dev Matched","Dev Mismatched"],"first_row_in_archive_order":{"model":"Turing NLR v5 XXL 5.4B (fine-tuned)","paper_title":null,"paper_url":null,"paper_date":"","arxiv_id":null,"code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-qnli","slug":"natural-language-inference-on-qnli","dataset":"QNLI","dataset_url":"/dataset/qnli","rows_in_archive":43,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ALBERT","paper_title":"ALBERT: A Lite BERT for Self-supervised Learning of Language Representations","paper_url":"/paper/albert-a-lite-bert-for-self-supervised","paper_date":"2019-09-26","arxiv_id":"1909.11942","code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"tensorflow/models","url":"https://github.com/tensorflow/models/tree/master/official/nlp/albert"},{"title":"PaddlePaddle/PaddleNLP","url":"https://github.com/PaddlePaddle/PaddleNLP/blob/develop/paddlenlp/transformers/albert/modeling.py"},{"title":"CLUEbenchmark/CLUE","url":"https://github.com/CLUEbenchmark/CLUE"},{"title":"brightmart/albert_zh","url":"https://github.com/brightmart/albert_zh"},{"title":"google-research/ALBERT","url":"https://github.com/google-research/ALBERT"},{"title":"Tencent/TurboTransformers","url":"https://github.com/Tencent/TurboTransformers"},{"title":"kpe/bert-for-tf2","url":"https://github.com/kpe/bert-for-tf2"},{"title":"lonePatient/albert_pytorch","url":"https://github.com/lonePatient/albert_pytorch"},{"title":"plkmo/BERT-Relation-Extraction","url":"https://github.com/plkmo/BERT-Relation-Extraction"},{"title":"facebookresearch/anli","url":"https://github.com/facebookresearch/anli"},{"title":"lucidrains/routing-transformer","url":"https://github.com/lucidrains/routing-transformer"},{"title":"lucidrains/sinkhorn-transformer","url":"https://github.com/lucidrains/sinkhorn-transformer"},{"title":"graykode/ALBERT-Pytorch","url":"https://github.com/graykode/ALBERT-Pytorch"},{"title":"kamalkraj/ALBERT-TF2.0","url":"https://github.com/kamalkraj/ALBERT-TF2.0"},{"title":"lucidrains/compressive-transformer-pytorch","url":"https://github.com/lucidrains/compressive-transformer-pytorch"},{"title":"epfml/collaborative-attention","url":"https://github.com/epfml/collaborative-attention"},{"title":"Sanyuan-Chen/RecAdam","url":"https://github.com/Sanyuan-Chen/RecAdam"},{"title":"mindspore-lab/mindnlp","url":"https://github.com/mindspore-lab/mindnlp/blob/master/mindnlp/models/albert/albert.py"},{"title":"benywon/ReCO","url":"https://github.com/benywon/ReCO"},{"title":"KnightZhang625/BERT_TF","url":"https://github.com/KnightZhang625/BERT_TF"},{"title":"AdamStein97/Semi-Supervised-BERT-NER","url":"https://github.com/AdamStein97/Semi-Supervised-BERT-NER"},{"title":"mandubian/codenets","url":"https://github.com/mandubian/codenets"},{"title":"codegram/calbert","url":"https://github.com/codegram/calbert"},{"title":"yahah100/text_summarization","url":"https://github.com/yahah100/text_summarization"},{"title":"Soikonomou/albert_final_infer12","url":"https://github.com/Soikonomou/albert_final_infer12"},{"title":"xinyooo/ALBERT4Rec","url":"https://github.com/xinyooo/ALBERT4Rec"},{"title":"rajatgermany/qa-nlp","url":"https://github.com/rajatgermany/qa-nlp"},{"title":"Soikonomou/bert_new","url":"https://github.com/Soikonomou/bert_new"},{"title":"Soikonomou/bert_new_new","url":"https://github.com/Soikonomou/bert_new_new"},{"title":"Soikonomou/albert_final","url":"https://github.com/Soikonomou/albert_final"},{"title":"Soikonomou/albert_final_infer8","url":"https://github.com/Soikonomou/albert_final_infer8"},{"title":"benywon/Chinese-GPT-2","url":"https://github.com/benywon/Chinese-GPT-2"},{"title":"jpablou/Matching-The-Blanks-Ths","url":"https://github.com/jpablou/Matching-The-Blanks-Ths"},{"title":"vvvm23/albert","url":"https://github.com/vvvm23/albert"},{"title":"common-english/bert-all","url":"https://github.com/common-english/bert-all"},{"title":"AhmedYounes94/Semi-supervised-BERT-NER","url":"https://github.com/AhmedYounes94/Semi-supervised-BERT-NER"},{"title":"appcoreopc/berty","url":"https://github.com/appcoreopc/berty"},{"title":"lyqcom/albert","url":"https://github.com/lyqcom/albert"},{"title":"mtzcorporations/nlp_teamjodka","url":"https://github.com/mtzcorporations/nlp_teamjodka"},{"title":"cui0523/Code6","url":"https://github.com/cui0523/Code6/tree/main/albert"},{"title":"cypressd1999/FYP_2021","url":"https://github.com/cypressd1999/FYP_2021"},{"title":"pwc-1/Paper-8","url":"https://github.com/pwc-1/Paper-8/tree/main/albert"},{"title":"Mind23-2/MindCode-131","url":"https://github.com/Mind23-2/MindCode-131"},{"title":"2023-MindSpore-1/ms-code-219","url":"https://github.com/2023-MindSpore-1/ms-code-219/tree/main/albert"},{"title":"hieudepchai/BERT_IE","url":"https://github.com/hieudepchai/BERT_IE"},{"title":"DataScienceNigeria/ALBERT-for-Natural-Language-Processing","url":"https://github.com/DataScienceNigeria/ALBERT-for-Natural-Language-Processing"},{"title":"MindSpore-paper-code-3/code9","url":"https://github.com/MindSpore-paper-code-3/code9/tree/main/albert"}],"syntology":{"n":126,"n_ran":68,"n_unverified":58,"n_pointer_only":22}}},{"leaderboard":"/sota/natural-language-inference-on-anli-test","slug":"natural-language-inference-on-anli-test","dataset":"ANLI test","dataset_url":"/dataset/anli","rows_in_archive":25,"metrics":["A1","A2","A3"],"first_row_in_archive_order":{"model":"T5-3B (explanation prompting)","paper_title":"Prompting for explanations improves Adversarial NLI. Is this true? {Yes} it is {true} because {it weakens superficial cues}","paper_url":"/paper/prompting-for-explanations-improves","paper_date":"2023-05-01","arxiv_id":null,"code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-wnli","slug":"natural-language-inference-on-wnli","dataset":"WNLI","dataset_url":"/dataset/wnli","rows_in_archive":23,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Turing NLR v5 XXL 5.4B (fine-tuned)","paper_title":null,"paper_url":null,"paper_date":"","arxiv_id":null,"code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-lidirus","slug":"natural-language-inference-on-lidirus","dataset":"LiDiRus","dataset_url":"/dataset/lidirus","rows_in_archive":22,"metrics":["MCC"],"first_row_in_archive_order":{"model":"Human Benchmark","paper_title":"RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark","paper_url":"/paper/russiansuperglue-a-russian-language","paper_date":"2020-10-29","arxiv_id":"2010.15925","code_links":[{"title":"RussianNLP/RussianSuperGLUE","url":"https://github.com/RussianNLP/RussianSuperGLUE"},{"title":"RussianNLP/MOROCCO","url":"https://github.com/RussianNLP/MOROCCO"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-rcb","slug":"natural-language-inference-on-rcb","dataset":"RCB","dataset_url":"/dataset/rcb","rows_in_archive":22,"metrics":["Average F1","Accuracy"],"first_row_in_archive_order":{"model":"Human Benchmark","paper_title":"RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark","paper_url":"/paper/russiansuperglue-a-russian-language","paper_date":"2020-10-29","arxiv_id":"2010.15925","code_links":[{"title":"RussianNLP/RussianSuperGLUE","url":"https://github.com/RussianNLP/RussianSuperGLUE"},{"title":"RussianNLP/MOROCCO","url":"https://github.com/RussianNLP/MOROCCO"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-terra","slug":"natural-language-inference-on-terra","dataset":"TERRa","dataset_url":"/dataset/terra","rows_in_archive":22,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Human Benchmark","paper_title":"RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark","paper_url":"/paper/russiansuperglue-a-russian-language","paper_date":"2020-10-29","arxiv_id":"2010.15925","code_links":[{"title":"RussianNLP/RussianSuperGLUE","url":"https://github.com/RussianNLP/RussianSuperGLUE"},{"title":"RussianNLP/MOROCCO","url":"https://github.com/RussianNLP/MOROCCO"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-commitmentbank","slug":"natural-language-inference-on-commitmentbank","dataset":"CommitmentBank","dataset_url":"/dataset/commitmentbank","rows_in_archive":20,"metrics":["Accuracy","F1"],"first_row_in_archive_order":{"model":"PaLM 540B (finetuned)","paper_title":"PaLM: Scaling Language Modeling with Pathways","paper_url":"/paper/palm-scaling-language-modeling-with-pathways-1","paper_date":"2022-04-05","arxiv_id":"2204.02311","code_links":[{"title":"lucidrains/CoCa-pytorch","url":"https://github.com/lucidrains/CoCa-pytorch"},{"title":"lucidrains/PaLM-pytorch","url":"https://github.com/lucidrains/PaLM-pytorch"},{"title":"google/paxml","url":"https://github.com/google/paxml"},{"title":"foundation-model-stack/fms-fsdp","url":"https://github.com/foundation-model-stack/fms-fsdp"},{"title":"lucidrains/PaLM-jax","url":"https://github.com/lucidrains/PaLM-jax"},{"title":"chrisociepa/allamo","url":"https://github.com/chrisociepa/allamo"},{"title":"conceptofmind/PaLM-flax","url":"https://github.com/conceptofmind/PaLM-flax"}],"syntology":{"n":37,"n_ran":32,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-scitail","slug":"natural-language-inference-on-scitail","dataset":"SciTail","dataset_url":"/dataset/scitail","rows_in_archive":13,"metrics":["Accuracy","Dev Accuracy","% Dev Accuracy","% Test Accuracy"],"first_row_in_archive_order":{"model":"CA-MTL","paper_title":"Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data","paper_url":"/paper/conditionally-adaptive-multi-task-learning","paper_date":"2020-09-19","arxiv_id":"2009.09139","code_links":[{"title":"CAMTL/CA-MTL","url":"https://github.com/CAMTL/CA-MTL"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-farstail","slug":"natural-language-inference-on-farstail","dataset":"FarsTail","dataset_url":"/dataset/farstail","rows_in_archive":10,"metrics":["% Test Accuracy"],"first_row_in_archive_order":{"model":"mBERT","paper_title":"FarsTail: A Persian Natural Language Inference Dataset","paper_url":"/paper/farstail-a-persian-natural-language-inference","paper_date":"2020-09-18","arxiv_id":"2009.08820","code_links":[{"title":"dml-qom/FarsTail","url":"https://github.com/dml-qom/FarsTail"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-multinli-dev","slug":"natural-language-inference-on-multinli-dev","dataset":"MultiNLI Dev","dataset_url":"/dataset/multinli","rows_in_archive":10,"metrics":["Matched","Mismatched"],"first_row_in_archive_order":{"model":"TinyBERT-6 67M","paper_title":"TinyBERT: Distilling BERT for Natural Language Understanding","paper_url":"/paper/190910351","paper_date":"2019-09-23","arxiv_id":"1909.10351","code_links":[{"title":"PaddlePaddle/PaddleNLP","url":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/paddlenlp/transformers/tinybert"},{"title":"huawei-noah/Pretrained-Language-Model","url":"https://github.com/huawei-noah/Pretrained-Language-Model/tree/master/TinyBERT"},{"title":"mindspore-ai/models","url":"https://github.com/mindspore-ai/models/tree/master/official/nlp/tinybert"},{"title":"mkavim/finetune_bert","url":"https://github.com/mkavim/finetune_bert"},{"title":"MindCode-4/code-5","url":"https://github.com/MindCode-4/code-5/tree/main/tinybert"},{"title":"pwc-1/Paper-9","url":"https://github.com/pwc-1/Paper-9/tree/main/1/tinybert"},{"title":"xiaolilaoli/tiny_bert_ms","url":"https://github.com/xiaolilaoli/tiny_bert_ms"},{"title":"pwc-1/Paper-10","url":"https://github.com/pwc-1/Paper-10/tree/main/tinybert"},{"title":"graison-thomas/TinyFinBERT","url":"https://github.com/graison-thomas/TinyFinBERT"},{"title":"2023-MindSpore-1/ms-code-166","url":"https://github.com/2023-MindSpore-1/ms-code-166"}],"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":4}}},{"leaderboard":"/sota/natural-language-inference-on-mednli","slug":"natural-language-inference-on-mednli","dataset":"MedNLI","dataset_url":"/dataset/mednli","rows_in_archive":7,"metrics":["Accuracy","Params (M)"],"first_row_in_archive_order":{"model":"ClinicalMosaic","paper_title":"Patient Trajectory Prediction: Integrating Clinical Notes with Transformers","paper_url":"/paper/patient-trajectory-prediction-integrating","paper_date":"2025-02-25","arxiv_id":"2502.18009","code_links":[{"title":"MostHumble/PatientTrajectoryForecasting","url":"https://github.com/MostHumble/PatientTrajectoryForecasting"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-xnli-french","slug":"natural-language-inference-on-xnli-french","dataset":"XNLI French","dataset_url":"/dataset/xnli","rows_in_archive":6,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"CamemBERT (large)","paper_title":"CamemBERT: a Tasty French Language Model","paper_url":"/paper/camembert-a-tasty-french-language-model","paper_date":"2019-11-10","arxiv_id":"1911.03894","code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"Karthik-Bhaskar/Context-Based-Question-Answering","url":"https://github.com/Karthik-Bhaskar/Context-Based-Question-Answering"},{"title":"anaishoareau/french_preprocessing","url":"https://github.com/anaishoareau/french_preprocessing"},{"title":"bourrel/French-News-Clustering","url":"https://github.com/bourrel/French-News-Clustering"},{"title":"2024-MindSpore-1/Code2","url":"https://github.com/2024-MindSpore-1/Code2/tree/main/model-1/camembert"},{"title":"hbaflast/bert-sentiment-analysis-pytorch","url":"https://github.com/hbaflast/bert-sentiment-analysis-pytorch"},{"title":"hbaflast/bert-sentiment-analysis-tensorflow","url":"https://github.com/hbaflast/bert-sentiment-analysis-tensorflow"},{"title":"pwc-1/Paper-8","url":"https://github.com/pwc-1/Paper-8/tree/main/camembert"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-e-snli","slug":"natural-language-inference-on-e-snli","dataset":"e-SNLI","dataset_url":"/dataset/e-snli","rows_in_archive":3,"metrics":["BLEU","Accuracy"],"first_row_in_archive_order":{"model":"ExplainThenPredictAttention (e-InferSent Bi-LSTM + Attention)","paper_title":"e-SNLI: Natural Language Inference with Natural Language Explanations","paper_url":"/paper/e-snli-natural-language-inference-with","paper_date":"2018-12-04","arxiv_id":"1812.01193","code_links":[{"title":"OanaMariaCamburu/e-SNLI","url":"https://github.com/OanaMariaCamburu/e-SNLI"},{"title":"qtli/eib","url":"https://github.com/qtli/eib"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-v-snli","slug":"natural-language-inference-on-v-snli","dataset":"V-SNLI","dataset_url":null,"rows_in_archive":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"MMBT","paper_title":"Supervised Multimodal Bitransformers for Classifying Images and Text","paper_url":"/paper/supervised-multimodal-bitransformers-for","paper_date":"2019-09-06","arxiv_id":"1909.02950","code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"ThilinaRajapakse/simpletransformers","url":"https://github.com/ThilinaRajapakse/simpletransformers"},{"title":"facebookresearch/mmbt","url":"https://github.com/facebookresearch/mmbt"},{"title":"IsaacRodgz/multimodal-transformers-movies","url":"https://github.com/IsaacRodgz/multimodal-transformers-movies"},{"title":"IsaacRodgz/mmbt_experiments","url":"https://github.com/IsaacRodgz/mmbt_experiments"},{"title":"adriangrepo/mmbt_lightning","url":"https://github.com/adriangrepo/mmbt_lightning"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-xnli-chinese","slug":"natural-language-inference-on-xnli-chinese","dataset":"XNLI Chinese Dev","dataset_url":"/dataset/xnli","rows_in_archive":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ERNIE 2.0 Large","paper_title":"ERNIE 2.0: A Continual Pre-training Framework for Language Understanding","paper_url":"/paper/ernie-20-a-continual-pre-training-framework","paper_date":"2019-07-29","arxiv_id":"1907.12412","code_links":[{"title":"PaddlePaddle/PaddleNLP","url":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0"},{"title":"PaddlePaddle/ERNIE","url":"https://github.com/PaddlePaddle/ERNIE"},{"title":"DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc.","url":"https://github.com/DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc."}],"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":1}}},{"leaderboard":"/sota/natural-language-inference-on-xnli-chinese-1","slug":"natural-language-inference-on-xnli-chinese-1","dataset":"XNLI Chinese","dataset_url":"/dataset/xnli","rows_in_archive":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ERNIE 2.0 Large","paper_title":"ERNIE 2.0: A Continual Pre-training Framework for Language Understanding","paper_url":"/paper/ernie-20-a-continual-pre-training-framework","paper_date":"2019-07-29","arxiv_id":"1907.12412","code_links":[{"title":"PaddlePaddle/PaddleNLP","url":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0"},{"title":"PaddlePaddle/ERNIE","url":"https://github.com/PaddlePaddle/ERNIE"},{"title":"DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc.","url":"https://github.com/DataScienceNigeria/ERNIE-2.0-from-Baidu-Inc."}],"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":1}}},{"leaderboard":"/sota/natural-language-inference-on-jampatoisnli","slug":"natural-language-inference-on-jampatoisnli","dataset":"JamPatoisNLI","dataset_url":"/dataset/jampatoisnli","rows_in_archive":2,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"roberta-unfrozen","paper_title":"JamPatoisNLI: A Jamaican Patois Natural Language Inference Dataset","paper_url":"/paper/jampatoisnli-a-jamaican-patois-natural","paper_date":"2022-12-07","arxiv_id":"2212.03419","code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-ax","slug":"natural-language-inference-on-ax","dataset":"AX","dataset_url":null,"rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"T5","paper_title":"SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization","paper_url":"/paper/smart-robust-and-efficient-fine-tuning-for","paper_date":"2019-11-08","arxiv_id":"1911.03437","code_links":[{"title":"namisan/mt-dnn","url":"https://github.com/namisan/mt-dnn"},{"title":"microsoft/MT-DNN","url":"https://github.com/microsoft/MT-DNN"},{"title":"archinetai/smart-pytorch","url":"https://github.com/archinetai/smart-pytorch"},{"title":"archinetai/vat-pytorch","url":"https://github.com/archinetai/vat-pytorch"},{"title":"cliang1453/camero","url":"https://github.com/cliang1453/camero"},{"title":"chunhuililili/mt_dnn","url":"https://github.com/chunhuililili/mt_dnn"}],"syntology":{"n":8,"n_ran":8,"n_unverified":0,"n_pointer_only":1}}},{"leaderboard":"/sota/natural-language-inference-on-bionli","slug":"natural-language-inference-on-bionli","dataset":"BioNLI","dataset_url":"/dataset/bionli","rows_in_archive":1,"metrics":["Macro F1"],"first_row_in_archive_order":{"model":"BioLinkBert","paper_title":"BioNLI: Generating a Biomedical NLI Dataset Using Lexico-semantic Constraints for Adversarial Examples","paper_url":"/paper/bionli-generating-a-biomedical-nli-dataset","paper_date":"2022-10-26","arxiv_id":"2210.14814","code_links":[{"title":"StonyBrookNLP/BioNLI","url":"https://github.com/StonyBrookNLP/BioNLI"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-hans","slug":"natural-language-inference-on-hans","dataset":"HANS","dataset_url":"/dataset/hans","rows_in_archive":1,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"Roberta-large","paper_title":"Generating Data to Mitigate Spurious Correlations in Natural Language Inference Datasets","paper_url":"/paper/generating-data-to-mitigate-spurious-1","paper_date":"2022-03-24","arxiv_id":"2203.12942","code_links":[{"title":"jimmycode/gen-debiased-nli","url":"https://github.com/jimmycode/gen-debiased-nli"}],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-kuake-qqr","slug":"natural-language-inference-on-kuake-qqr","dataset":"KUAKE-QQR","dataset_url":"/dataset/kuake-qqr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"BERT-base","paper_title":"CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark","paper_url":"/paper/cblue-a-chinese-biomedical-language","paper_date":"2021-06-15","arxiv_id":"2106.08087","code_links":[{"title":"cbluebenchmark/cblue","url":"https://github.com/cbluebenchmark/cblue"},{"title":"freedomintelligence/sdak","url":"https://github.com/freedomintelligence/sdak"}],"syntology":{"n":16,"n_ran":4,"n_unverified":12,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-kuake-qtr","slug":"natural-language-inference-on-kuake-qtr","dataset":"KUAKE-QTR","dataset_url":"/dataset/kuake-qtr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"MacBERT-large","paper_title":"CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark","paper_url":"/paper/cblue-a-chinese-biomedical-language","paper_date":"2021-06-15","arxiv_id":"2106.08087","code_links":[{"title":"cbluebenchmark/cblue","url":"https://github.com/cbluebenchmark/cblue"},{"title":"freedomintelligence/sdak","url":"https://github.com/freedomintelligence/sdak"}],"syntology":{"n":16,"n_ran":4,"n_unverified":12,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-med","slug":"natural-language-inference-on-med","dataset":"MED","dataset_url":"/dataset/med","rows_in_archive":1,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"NeuralLog","paper_title":"NeuralLog: Natural Language Inference with Joint Neural and Logical Reasoning","paper_url":"/paper/neurallog-natural-language-inference-with","paper_date":"2021-05-29","arxiv_id":"2105.14167","code_links":[{"title":"eric11eca/NeuralLog","url":"https://github.com/eric11eca/NeuralLog"}],"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-mnli-snli-anli","slug":"natural-language-inference-on-mnli-snli-anli","dataset":"MNLI + SNLI + ANLI + FEVER","dataset_url":null,"rows_in_archive":1,"metrics":["% Dev Accuracy","% Test Accuracy"],"first_row_in_archive_order":{"model":"SMARTRoBERTa-LARGE","paper_title":"SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization","paper_url":"/paper/smart-robust-and-efficient-fine-tuning-for","paper_date":"2019-11-08","arxiv_id":"1911.03437","code_links":[{"title":"namisan/mt-dnn","url":"https://github.com/namisan/mt-dnn"},{"title":"microsoft/MT-DNN","url":"https://github.com/microsoft/MT-DNN"},{"title":"archinetai/smart-pytorch","url":"https://github.com/archinetai/smart-pytorch"},{"title":"archinetai/vat-pytorch","url":"https://github.com/archinetai/vat-pytorch"},{"title":"cliang1453/camero","url":"https://github.com/cliang1453/camero"},{"title":"chunhuililili/mt_dnn","url":"https://github.com/chunhuililili/mt_dnn"}],"syntology":{"n":8,"n_ran":8,"n_unverified":0,"n_pointer_only":1}}},{"leaderboard":"/sota/natural-language-inference-on-mrpc","slug":"natural-language-inference-on-mrpc","dataset":"MRPC","dataset_url":"/dataset/mrpc","rows_in_archive":1,"metrics":["Acc"],"first_row_in_archive_order":{"model":"DeBERTaV3large","paper_title":"DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing","paper_url":"/paper/debertav3-improving-deberta-using-electra","paper_date":"2021-11-18","arxiv_id":"2111.09543","code_links":[{"title":"microsoft/DeBERTa","url":"https://github.com/microsoft/DeBERTa"},{"title":"dashenzi721/hra","url":"https://github.com/dashenzi721/hra"},{"title":"stareru/csqa_debertav3","url":"https://github.com/stareru/csqa_debertav3"}],"syntology":{"n":7,"n_ran":6,"n_unverified":1,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-probability","slug":"natural-language-inference-on-probability","dataset":"Probability words NLI","dataset_url":"/dataset/probability-words-nli","rows_in_archive":1,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"roberta-base-mnli","paper_title":"Probing neural language models for understanding of words of estimative probability","paper_url":"/paper/probing-neural-language-models-for","paper_date":"2022-11-07","arxiv_id":"2211.03358","code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-quora-question","slug":"natural-language-inference-on-quora-question","dataset":"Quora Question Pairs","dataset_url":"/dataset/quora-question-pairs","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"aESIM","paper_title":"Attention Boosted Sequential Inference Model","paper_url":"/paper/attention-boosted-sequential-inference-model","paper_date":"2018-12-05","arxiv_id":"1812.01840","code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-sick","slug":"natural-language-inference-on-sick","dataset":"SICK","dataset_url":"/dataset/sick","rows_in_archive":1,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"NeuralLog","paper_title":"NeuralLog: Natural Language Inference with Joint Neural and Logical Reasoning","paper_url":"/paper/neurallog-natural-language-inference-with","paper_date":"2021-05-29","arxiv_id":"2105.14167","code_links":[{"title":"eric11eca/NeuralLog","url":"https://github.com/eric11eca/NeuralLog"}],"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}}},{"leaderboard":"/sota/natural-language-inference-on-tabfact","slug":"natural-language-inference-on-tabfact","dataset":"TabFact","dataset_url":"/dataset/tabfact","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ChatGPT 3.5 SpatialFormat","paper_title":"LAPDoc: Layout-Aware Prompting for Documents","paper_url":"/paper/lapdoc-layout-aware-prompting-for-documents","paper_date":"2024-02-15","arxiv_id":"2402.09841","code_links":[],"syntology":null}},{"leaderboard":"/sota/natural-language-inference-on-xwino","slug":"natural-language-inference-on-xwino","dataset":"XWINO","dataset_url":"/dataset/xwino","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"mGPT","paper_title":"mGPT: Few-Shot Learners Go Multilingual","paper_url":"/paper/mgpt-few-shot-learners-go-multilingual","paper_date":"2022-04-15","arxiv_id":"2204.07580","code_links":[{"title":"ai-forever/mgpt","url":"https://github.com/ai-forever/mgpt"}],"syntology":null}},{"leaderboard":null,"slug":"natural-language-inference-on-anli","dataset":"ANLI","dataset_url":"/dataset/anli","rows_in_archive":0,"metrics":["Accuracy","F1 Macro","F1 Micro","F1 Weighted","Precision Macro","Precision Micro","Precision Weighted","Recall Macro","Recall Micro","Recall Weighted","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"natural-language-inference-on-anli-r3","dataset":"ANLI-r3","dataset_url":"/dataset/anli","rows_in_archive":0,"metrics":["Accuracy"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"natural-language-inference-on-glue","dataset":"GLUE","dataset_url":"/dataset/glue","rows_in_archive":0,"metrics":["Accuracy","F1 Macro","F1 Micro","F1 Weighted","Precision Macro","Precision Micro","Precision Weighted","Recall Macro","Recall Micro","Recall Weighted","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"natural-language-inference-on-multi-nli","dataset":"multi_nli","dataset_url":"/dataset/multinli","rows_in_archive":0,"metrics":["Accuracy","F1 Macro","F1 Micro","F1 Weighted","Precision Macro","Precision Micro","Precision Weighted","Recall Macro","Recall Micro","Recall Weighted","loss"],"first_row_in_archive_order":null}],"datasets":[{"url":"/dataset/glue","name":"GLUE","full_name":"General Language Understanding Evaluation benchmark","num_papers_in_archive":3197},{"url":"/dataset/multinli","name":"MultiNLI","full_name":"Multi-Genre Natural Language Inference","num_papers_in_archive":1830},{"url":"/dataset/snli","name":"SNLI","full_name":"Stanford Natural Language Inference","num_papers_in_archive":1311},{"url":"/dataset/qnli","name":"QNLI","full_name":"Question-answering NLI","num_papers_in_archive":1234},{"url":"/dataset/mrpc","name":"MRPC","full_name":"Microsoft Research Paraphrase Corpus","num_papers_in_archive":786},{"url":"/dataset/winogrande","name":"WinoGrande","full_name":"","num_papers_in_archive":703},{"url":"/dataset/xnli","name":"XNLI","full_name":"Cross-lingual Natural Language Inference","num_papers_in_archive":349},{"url":"/dataset/sick","name":"SICK","full_name":"Sentences Involving Compositional Knowledge","num_papers_in_archive":348},{"url":"/dataset/anli","name":"ANLI","full_name":"Adversarial NLI","num_papers_in_archive":287},{"url":"/dataset/paws","name":"PAWS","full_name":"Paraphrase Adversaries from Word Scrambling","num_papers_in_archive":159},{"url":"/dataset/e-snli","name":"e-SNLI","full_name":"","num_papers_in_archive":139},{"url":"/dataset/tabfact","name":"TabFact","full_name":"TabFact","num_papers_in_archive":129},{"url":"/dataset/snli-ve","name":"SNLI-VE","full_name":null,"num_papers_in_archive":117},{"url":"/dataset/gap-coreference-dataset","name":"GAP Coreference Dataset","full_name":"","num_papers_in_archive":104},{"url":"/dataset/listops","name":"ListOps","full_name":"ListOps","num_papers_in_archive":97},{"url":"/dataset/aqua-rat","name":"AQUA-RAT","full_name":"Algebra Question Answering with Rationales","num_papers_in_archive":64},{"url":"/dataset/rte","name":"RTE","full_name":"Recognizing Textual Entailment","num_papers_in_archive":56},{"url":"/dataset/quora-question-pairs","name":"Quora Question Pairs","full_name":"","num_papers_in_archive":55},{"url":"/dataset/ocnli","name":"OCNLI","full_name":"Original Chinese Natural Language Inference","num_papers_in_archive":44},{"url":"/dataset/scrolls","name":"SCROLLS","full_name":"Standardized CompaRison Over Long Language Sequences","num_papers_in_archive":42},{"url":"/dataset/break","name":"BREAK","full_name":"BREAK","num_papers_in_archive":39},{"url":"/dataset/medquad","name":"MedQuAD","full_name":"Medical Question Answering Dataset","num_papers_in_archive":32},{"url":"/dataset/contractnli","name":"ContractNLI","full_name":"","num_papers_in_archive":31},{"url":"/dataset/help","name":"HELP","full_name":"","num_papers_in_archive":30},{"url":"/dataset/chaosnli","name":"ChaosNLI","full_name":"","num_papers_in_archive":27},{"url":"/dataset/headqa","name":"HeadQA","full_name":"","num_papers_in_archive":23},{"url":"/dataset/xglue","name":"XGLUE","full_name":"","num_papers_in_archive":22},{"url":"/dataset/klue","name":"KLUE","full_name":"Korean Language Understanding Evaluation","num_papers_in_archive":21},{"url":"/dataset/quarel","name":"QuaRel","full_name":"QuaRel","num_papers_in_archive":20},{"url":"/dataset/torque","name":"Torque","full_name":"","num_papers_in_archive":19},{"url":"/dataset/infotabs","name":"InfoTabS","full_name":"","num_papers_in_archive":18},{"url":"/dataset/kornli","name":"KorNLI","full_name":null,"num_papers_in_archive":18},{"url":"/dataset/med","name":"MED","full_name":"Monotonicity Entailment Dataset","num_papers_in_archive":18},{"url":"/dataset/indicglue","name":"IndicGLUE","full_name":"Indic General Language Understanding Evaluation Benchmark","num_papers_in_archive":16},{"url":"/dataset/wnli","name":"WNLI","full_name":"Winograd NLI","num_papers_in_archive":14},{"url":"/dataset/korsts","name":"KorSTS","full_name":"","num_papers_in_archive":13},{"url":"/dataset/story-commonsense","name":"Story Commonsense","full_name":"","num_papers_in_archive":13},{"url":"/dataset/flue-french-language-understanding-evaluation","name":"FLUE","full_name":"French Language Understanding Evaluation","num_papers_in_archive":12},{"url":"/dataset/taxinli","name":"TaxiNLI","full_name":"","num_papers_in_archive":11},{"url":"/dataset/commitmentbank","name":"CommitmentBank","full_name":"","num_papers_in_archive":10},{"url":"/dataset/nli4ct","name":"NLI4CT","full_name":"","num_papers_in_archive":10},{"url":"/dataset/art-dataset","name":"ART Dataset","full_name":"Abductive Reasoning in narrative Text","num_papers_in_archive":9},{"url":"/dataset/farstail","name":"FarsTail","full_name":"","num_papers_in_archive":9},{"url":"/dataset/mednli","name":"MedNLI","full_name":"Medical Natural Language Inference","num_papers_in_archive":9},{"url":"/dataset/scitail","name":"SciTail","full_name":"","num_papers_in_archive":9},{"url":"/dataset/selqa","name":"SelQA","full_name":"","num_papers_in_archive":9},{"url":"/dataset/docnli","name":"DocNLI","full_name":"","num_papers_in_archive":8},{"url":"/dataset/danetqa","name":"DaNetQA","full_name":"Yes/no Question Answering Dataset for the Russian","num_papers_in_archive":7},{"url":"/dataset/marc-ja","name":"JGLUE","full_name":"","num_papers_in_archive":7},{"url":"/dataset/kuake-qqr","name":"KUAKE-QQR","full_name":"Query-Query Relevance Dataset","num_papers_in_archive":7},{"url":"/dataset/kuake-qtr","name":"KUAKE-QTR","full_name":"Query-Title Relevance Dataset","num_papers_in_archive":7},{"url":"/dataset/terra","name":"TERRa","full_name":"Textual Entailment Recognition for Russian","num_papers_in_archive":7},{"url":"/dataset/imppres","name":"IMPPRES","full_name":"","num_papers_in_archive":6},{"url":"/dataset/rcb","name":"RCB","full_name":"Russian Commitment Bank","num_papers_in_archive":6},{"url":"/dataset/bionli","name":"BioNLI","full_name":"Biomedical Natural Language Inference","num_papers_in_archive":5},{"url":"/dataset/lidirus","name":"LiDiRus","full_name":"Linguistic Diagnostic for Russian","num_papers_in_archive":5},{"url":"/dataset/rusentrel","name":"RuSentRel","full_name":"","num_papers_in_archive":5},{"url":"/dataset/indonli","name":"IndoNLI","full_name":"","num_papers_in_archive":4},{"url":"/dataset/mindgames","name":"Mindgames","full_name":"","num_papers_in_archive":4},{"url":"/dataset/xwino","name":"XWINO","full_name":"","num_papers_in_archive":4},{"url":"/dataset/esxnli","name":"esXNLI","full_name":null,"num_papers_in_archive":3},{"url":"/dataset/janli","name":"JaNLI","full_name":"Japanese Adversarial Natural Language Inference","num_papers_in_archive":3},{"url":"/dataset/justice","name":"JUSTICE","full_name":"JUSTICE: A Dataset for Supreme Court’s Judgment Prediction","num_papers_in_archive":3},{"url":"/dataset/pars-absa","name":"Pars-ABSA","full_name":"","num_papers_in_archive":3},{"url":"/dataset/tasksource","name":"Tasksource","full_name":"","num_papers_in_archive":3},{"url":"/dataset/wili-2018","name":"WiLI-2018","full_name":"","num_papers_in_archive":3},{"url":"/dataset/arnli","name":"ArNLI","full_name":"","num_papers_in_archive":2},{"url":"/dataset/newsph-nli","name":"NewsPH-NLI","full_name":"","num_papers_in_archive":2},{"url":"/dataset/nli-tr","name":"NLI-TR","full_name":"Natural Language Inference in Turkish","num_papers_in_archive":2},{"url":"/dataset/propsegment","name":"PropSegmEnt","full_name":"","num_papers_in_archive":2},{"url":"/dataset/candor-corpus-1","name":"CANDOR Corpus","full_name":"CANDOR = Conversation: A Naturalistic Dataset of Online Recordings","num_papers_in_archive":1},{"url":"/dataset/distnli","name":"DistNLI","full_name":"","num_papers_in_archive":1},{"url":"/dataset/gd-nli","name":"GD-NLI","full_name":"Generated Debiased NLI Datasets","num_papers_in_archive":1},{"url":"/dataset/gigaword-entailment","name":"Gigaword Entailment","full_name":null,"num_papers_in_archive":1},{"url":"/dataset/gqnli-fr","name":"GQNLI-FR","full_name":"","num_papers_in_archive":1},{"url":"/dataset/hans","name":"HANS","full_name":"Heuristic Analysis for NLI Systems","num_papers_in_archive":1},{"url":"/dataset/jampatoisnli","name":"JamPatoisNLI","full_name":"","num_papers_in_archive":1},{"url":"/dataset/nli4wills-corpus","name":"NLI4Wills Corpus","full_name":"","num_papers_in_archive":1},{"url":"/dataset/probability-words-nli","name":"Probability words NLI","full_name":"Natural language inference with words estimative of probability (WEP)","num_papers_in_archive":1},{"url":"/dataset/probability-words-nli-1","name":"probability_words_nli","full_name":"","num_papers_in_archive":1},{"url":"/dataset/rte3-fr","name":"RTE3-FR","full_name":"","num_papers_in_archive":1}],"subtasks":[{"url":"/task/answer-generation","name":"Answer Generation"},{"url":"/task/cross-lingual-natural-language-inference","name":"Cross-Lingual Natural Language Inference"},{"url":"/task/visual-entailment","name":"Visual Entailment"}],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":821,"tagged_in_all":1961,"items":[{"url":"/paper/bert-pre-training-of-deep-bidirectional","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","date":"2018-10-11","arxiv_id":"1810.04805","repositories_listed":534,"syntology":{"n":659,"n_ran":208,"n_unverified":451,"n_pointer_only":149}},{"url":"/paper/language-models-are-few-shot-learners","title":"Language Models are Few-Shot Learners","date":"2020-05-28","arxiv_id":"2005.14165","repositories_listed":67,"syntology":{"n":65,"n_ran":41,"n_unverified":24,"n_pointer_only":4}},{"url":"/paper/roberta-a-robustly-optimized-bert-pretraining","title":"RoBERTa: A Robustly Optimized BERT Pretraining Approach","date":"2019-07-26","arxiv_id":"1907.11692","repositories_listed":67,"syntology":{"n":48,"n_ran":37,"n_unverified":11,"n_pointer_only":23}},{"url":"/paper/exploring-the-limits-of-transfer-learning","title":"Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer","date":"2019-10-23","arxiv_id":"1910.10683","repositories_listed":57,"syntology":{"n":31,"n_ran":21,"n_unverified":10,"n_pointer_only":0}},{"url":"/paper/a-structured-self-attentive-sentence","title":"A Structured Self-attentive Sentence Embedding","date":"2017-03-09","arxiv_id":"1703.03130","repositories_listed":52,"syntology":{"n":10,"n_ran":4,"n_unverified":6,"n_pointer_only":2}},{"url":"/paper/albert-a-lite-bert-for-self-supervised","title":"ALBERT: A Lite BERT for Self-supervised Learning of Language Representations","date":"2019-09-26","arxiv_id":"1909.11942","repositories_listed":48,"syntology":{"n":126,"n_ran":68,"n_unverified":58,"n_pointer_only":22}},{"url":"/paper/bart-denoising-sequence-to-sequence-pre","title":"BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension","date":"2019-10-29","arxiv_id":"1910.13461","repositories_listed":47,"syntology":{"n":53,"n_ran":39,"n_unverified":14,"n_pointer_only":7}},{"url":"/paper/deep-contextualized-word-representations","title":"Deep contextualized word representations","date":"2018-02-15","arxiv_id":"1802.05365","repositories_listed":46,"syntology":{"n":58,"n_ran":24,"n_unverified":34,"n_pointer_only":25}},{"url":"/paper/distilbert-a-distilled-version-of-bert","title":"DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter","date":"2019-10-02","arxiv_id":"1910.01108","repositories_listed":37,"syntology":{"n":27,"n_ran":20,"n_unverified":7,"n_pointer_only":0}},{"url":"/paper/xlnet-generalized-autoregressive-pretraining","title":"XLNet: Generalized Autoregressive Pretraining for Language Understanding","date":"2019-06-19","arxiv_id":"1906.08237","repositories_listed":27,"syntology":{"n":24,"n_ran":10,"n_unverified":14,"n_pointer_only":3}},{"url":"/paper/simcse-simple-contrastive-learning-of","title":"SimCSE: Simple Contrastive Learning of Sentence Embeddings","date":"2021-04-18","arxiv_id":"2104.08821","repositories_listed":23,"syntology":{"n":30,"n_ran":17,"n_unverified":13,"n_pointer_only":19}},{"url":"/paper/supervised-learning-of-universal-sentence","title":"Supervised Learning of Universal Sentence Representations from Natural Language Inference Data","date":"2017-05-05","arxiv_id":"1705.02364","repositories_listed":23,"syntology":{"n":7,"n_ran":6,"n_unverified":1,"n_pointer_only":7}},{"url":"/paper/pay-attention-to-mlps","title":"Pay Attention to MLPs","date":"2021-05-17","arxiv_id":"2105.08050","repositories_listed":20,"syntology":{"n":44,"n_ran":34,"n_unverified":10,"n_pointer_only":11}},{"url":"/paper/ernie-enhanced-representation-through","title":"ERNIE: Enhanced Representation through Knowledge Integration","date":"2019-04-19","arxiv_id":"1904.09223","repositories_listed":19,"syntology":{"n":7,"n_ran":6,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/rwkv-reinventing-rnns-for-the-transformer-era","title":"RWKV: Reinventing RNNs for the Transformer Era","date":"2023-05-22","arxiv_id":"2305.13048","repositories_listed":14,"syntology":{"n":11,"n_ran":6,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/big-bird-transformers-for-longer-sequences","title":"Big Bird: Transformers for Longer Sequences","date":"2020-07-28","arxiv_id":"2007.14062","repositories_listed":14,"syntology":{"n":15,"n_ran":10,"n_unverified":5,"n_pointer_only":11}},{"url":"/paper/deberta-decoding-enhanced-bert-with","title":"DeBERTa: Decoding-enhanced BERT with Disentangled Attention","date":"2020-06-05","arxiv_id":"2006.03654","repositories_listed":14,"syntology":{"n":13,"n_ran":4,"n_unverified":9,"n_pointer_only":3}},{"url":"/paper/massively-multilingual-sentence-embeddings","title":"Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond","date":"2018-12-26","arxiv_id":"1812.10464","repositories_listed":13,"syntology":{"n":10,"n_ran":4,"n_unverified":6,"n_pointer_only":4}},{"url":"/paper/improving-language-understanding-by","title":"Improving Language Understanding by Generative Pre-Training","date":"2018-06-11","arxiv_id":null,"repositories_listed":13,"syntology":null},{"url":"/paper/data2vec-a-general-framework-for-self-1","title":"data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language","date":"2022-02-07","arxiv_id":"2202.03555","repositories_listed":12,"syntology":{"n":6,"n_ran":4,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/fnet-mixing-tokens-with-fourier-transforms","title":"FNet: Mixing Tokens with Fourier Transforms","date":"2021-05-09","arxiv_id":"2105.03824","repositories_listed":12,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":1}},{"url":"/paper/enhanced-lstm-for-natural-language-inference","title":"Enhanced LSTM for Natural Language Inference","date":"2016-09-20","arxiv_id":"1609.06038","repositories_listed":12,"syntology":null},{"url":"/paper/glue-a-multi-task-benchmark-and-analysis","title":"GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding","date":"2018-04-20","arxiv_id":"1804.07461","repositories_listed":11,"syntology":{"n":14,"n_ran":5,"n_unverified":9,"n_pointer_only":1}},{"url":"/paper/senteval-an-evaluation-toolkit-for-universal","title":"SentEval: An Evaluation Toolkit for Universal Sentence Representations","date":"2018-03-14","arxiv_id":"1803.05449","repositories_listed":11,"syntology":{"n":8,"n_ran":6,"n_unverified":2,"n_pointer_only":6}},{"url":"/paper/from-softmax-to-sparsemax-a-sparse-model-of","title":"From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classification","date":"2016-02-05","arxiv_id":"1602.02068","repositories_listed":11,"syntology":{"n":8,"n_ran":8,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/new-datasets-for-automatic-detection-of","title":"New Datasets for Automatic Detection of Textual Entailment and of Contradictions between Sentences in French","date":"2024-05-23","arxiv_id":null,"repositories_listed":10,"syntology":null},{"url":"/paper/nystromformer-a-nystrom-based-algorithm-for","title":"Nyströmformer: A Nyström-Based Algorithm for Approximating Self-Attention","date":"2021-02-07","arxiv_id":"2102.03902","repositories_listed":10,"syntology":{"n":2,"n_ran":1,"n_unverified":1,"n_pointer_only":1}},{"url":"/paper/190910351","title":"TinyBERT: Distilling BERT for Natural Language Understanding","date":"2019-09-23","arxiv_id":"1909.10351","repositories_listed":10,"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":4}},{"url":"/paper/nezha-neural-contextualized-representation","title":"NEZHA: Neural Contextualized Representation for Chinese Language Understanding","date":"2019-08-31","arxiv_id":"1909.00204","repositories_listed":10,"syntology":null},{"url":"/paper/bilateral-multi-perspective-matching-for","title":"Bilateral Multi-Perspective Matching for Natural Language Sentences","date":"2017-02-13","arxiv_id":"1702.03814","repositories_listed":10,"syntology":{"n":4,"n_ran":1,"n_unverified":3,"n_pointer_only":1}}],"syntology_records":26,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":3,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}