{"url":"/dataset/qnli","name":"QNLI","full_name":"Question-answering NLI","description_markdown":"The **QNLI** (**Question-answering NLI**) dataset is a Natural Language Inference dataset automatically derived from the Stanford Question Answering Dataset v1.1 (SQuAD). SQuAD v1.1 consists of question-paragraph pairs, where one of the sentences in the paragraph (drawn from Wikipedia) contains the answer to the corresponding question (written by an annotator). The dataset was converted into sentence pair classification by forming a pair between each question and each sentence in the corresponding context, and filtering out pairs with low lexical overlap between the question and the context sentence. The task is to determine whether the context sentence contains the answer to the question. This modified version of the original task removes the requirement that the model select the exact answer, but also removes the simplifying assumptions that the answer is always present in the input and that lexical overlap is a reliable cue. The QNLI dataset is part of GLUE benchmark.\r\n\r\nSource: [https://arxiv.org/pdf/1804.07461.pdf](https://arxiv.org/pdf/1804.07461.pdf)","description_withheld":null,"homepage":"https://gluebenchmark.com/","introduced_date":"2019-01-01","introduced_date_note":null,"introduced_by":{"paper":"/paper/glue-a-multi-task-benchmark-and-analysis","title":"GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding","first_author":"Alex Wang","url":null},"license":{"name":"CC BY-SA 4.0","url":null},"modalities":[{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Natural Language Inference","url":"/task/natural-language-inference","datasets_with_task":"/datasets/task/natural-language-inference"},{"name":"Data-free Knowledge Distillation","url":"/task/data-free-knowledge-distillation","datasets_with_task":"/datasets/task/data-free-knowledge-distillation"},{"name":"Model Compression","url":"/task/model-compression","datasets_with_task":"/datasets/task/model-compression"},{"name":"Few-Shot NLI","url":"/task/few-shot-nli","datasets_with_task":"/datasets/task/few-shot-nli"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["QNLI","QNLI (8 training examples per class)","QNLI Dev"],"data_loaders":[],"num_papers_in_archive":1234,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/natural-language-inference-on-qnli","task":"Natural Language Inference","dataset_variant":"QNLI","rows":43,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ALBERT","paper":"/paper/albert-a-lite-bert-for-self-supervised","metrics":{"Accuracy":"99.2%"},"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"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/data-free-knowledge-distillation-on-qnli","task":"Data-free Knowledge Distillation","dataset_variant":"QNLI","rows":4,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"GOLD (T5-base)","paper":"/paper/gold-generalized-knowledge-distillation-via","metrics":{"Accuracy":"91.7"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/model-compression-on-qnli","task":"Model Compression","dataset_variant":"QNLI","rows":2,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"MobileBERT + 2bit-1dim model compression using DKM","paper":"/paper/r-2-range-regularization-for-model","metrics":{"Accuracy":"82.13"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/gold-generalized-knowledge-distillation-via","title":"GOLD: Generalized Knowledge Distillation via Out-of-Distribution-Guided Language Data Generation","date":"2024-03-28","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/prompt2model-generating-deployable-models","title":"Prompt2Model: Generating Deployable Models from Natural Language Instructions","date":"2023-08-23","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/lm-cppf-paraphrasing-guided-data-augmentation","title":"LM-CPPF: Paraphrasing-Guided Data Augmentation for Contrastive Prompt-Based Few-Shot Fine-Tuning","date":"2023-05-29","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/r-2-range-regularization-for-model","title":"R2 Loss: Range Restriction Loss for Model Compression and Quantization","date":"2023-03-14","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/progen-progressive-zero-shot-dataset","title":"ProGen: Progressive Zero-shot Dataset Generation via In-context Feedback","date":"2022-10-22","rows_on_this_dataset":1,"code_links":2,"syntology":null},{"paper":"/paper/llm-int8-8-bit-matrix-multiplication-for","title":"LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale","date":"2022-08-15","rows_on_this_dataset":1,"code_links":4,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":5,"samples_ran":4,"samples_unverified":1,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/adversarial-self-attention-for-language","title":"Adversarial Self-Attention for Language Understanding","date":"2022-06-25","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/zerogen-efficient-zero-shot-learning-via","title":"ZeroGen: Efficient Zero-shot Learning via Dataset Generation","date":"2022-02-16","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":7,"samples_ran":3,"samples_unverified":4,"pointer_only_for_licence":7,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/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","rows_on_this_dataset":1,"code_links":12,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":6,"samples_ran":4,"samples_unverified":2,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/debertav3-improving-deberta-using-electra","title":"DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing","date":"2021-11-18","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":7,"samples_ran":6,"samples_unverified":1,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/charformer-fast-character-transformers-via","title":"Charformer: Fast Character Transformers via Gradient-based Subword Tokenization","date":"2021-06-23","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":10,"samples_ran":7,"samples_unverified":3,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/fnet-mixing-tokens-with-fourier-transforms","title":"FNet: Mixing Tokens with Fourier Transforms","date":"2021-05-09","rows_on_this_dataset":1,"code_links":12,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":2,"samples_ran":2,"samples_unverified":0,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/entailment-as-few-shot-learner","title":"Entailment as Few-Shot Learner","date":"2021-04-29","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":3,"samples_ran":1,"samples_unverified":2,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/how-to-train-bert-with-an-academic-budget","title":"How to Train BERT with an Academic Budget","date":"2021-04-15","rows_on_this_dataset":1,"code_links":4,"syntology":null},{"paper":"/paper/nystromformer-a-nystrom-based-algorithm-for","title":"Nyströmformer: A Nyström-Based Algorithm for Approximating Self-Attention","date":"2021-02-07","rows_on_this_dataset":1,"code_links":10,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":2,"samples_ran":1,"samples_unverified":1,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/clear-contrastive-learning-for-sentence","title":"CLEAR: Contrastive Learning for Sentence Representation","date":"2020-12-31","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/informer-transformer-likes-informed-attention","title":"RealFormer: Transformer Likes Residual Attention","date":"2020-12-21","rows_on_this_dataset":1,"code_links":5,"syntology":null},{"paper":"/paper/a-statistical-framework-for-low-bitwidth","title":"A Statistical Framework for Low-bitwidth Training of Deep Neural Networks","date":"2020-10-27","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":4,"samples_ran":1,"samples_unverified":3,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/big-bird-transformers-for-longer-sequences","title":"Big Bird: Transformers for Longer Sequences","date":"2020-07-28","rows_on_this_dataset":1,"code_links":14,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":15,"samples_ran":10,"samples_unverified":5,"pointer_only_for_licence":11,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/squeezebert-what-can-computer-vision-teach","title":"SqueezeBERT: What can computer vision teach NLP about efficient neural networks?","date":"2020-06-19","rows_on_this_dataset":1,"code_links":6,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/deberta-decoding-enhanced-bert-with","title":"DeBERTa: Decoding-enhanced BERT with Disentangled Attention","date":"2020-06-05","rows_on_this_dataset":1,"code_links":14,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":13,"samples_ran":4,"samples_unverified":9,"pointer_only_for_licence":3,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/trans-blstm-transformer-with-bidirectional","title":"TRANS-BLSTM: Transformer with Bidirectional LSTM for Language Understanding","date":"2020-03-16","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/smart-robust-and-efficient-fine-tuning-for","title":"SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization","date":"2019-11-08","rows_on_this_dataset":4,"code_links":6,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":8,"samples_ran":8,"samples_unverified":0,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/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","rows_on_this_dataset":5,"code_links":57,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":31,"samples_ran":21,"samples_unverified":10,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/q8bert-quantized-8bit-bert","title":"Q8BERT: Quantized 8Bit BERT","date":"2019-10-14","rows_on_this_dataset":1,"code_links":5,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":11,"samples_ran":5,"samples_unverified":6,"pointer_only_for_licence":3,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/distilbert-a-distilled-version-of-bert","title":"DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter","date":"2019-10-02","rows_on_this_dataset":1,"code_links":37,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":27,"samples_ran":20,"samples_unverified":7,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/albert-a-lite-bert-for-self-supervised","title":"ALBERT: A Lite BERT for Self-supervised Learning of Language Representations","date":"2019-09-26","rows_on_this_dataset":1,"code_links":48,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":126,"samples_ran":68,"samples_unverified":58,"pointer_only_for_licence":22,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/190910351","title":"TinyBERT: Distilling BERT for Natural Language Understanding","date":"2019-09-23","rows_on_this_dataset":2,"code_links":10,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":4,"samples_ran":0,"samples_unverified":4,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/q-bert-hessian-based-ultra-low-precision","title":"Q-BERT: Hessian Based Ultra Low Precision Quantization of BERT","date":"2019-09-12","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/sensebert-driving-some-sense-into-bert","title":"SenseBERT: Driving Some Sense into BERT","date":"2019-08-15","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/structbert-incorporating-language-structures","title":"StructBERT: Incorporating Language Structures into Pre-training for Deep Language Understanding","date":"2019-08-13","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/ernie-20-a-continual-pre-training-framework","title":"ERNIE 2.0: A Continual Pre-training Framework for Language Understanding","date":"2019-07-29","rows_on_this_dataset":2,"code_links":3,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":1,"samples_ran":0,"samples_unverified":1,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/roberta-a-robustly-optimized-bert-pretraining","title":"RoBERTa: A Robustly Optimized BERT Pretraining Approach","date":"2019-07-26","rows_on_this_dataset":1,"code_links":67,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":48,"samples_ran":37,"samples_unverified":11,"pointer_only_for_licence":23,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/spanbert-improving-pre-training-by","title":"SpanBERT: Improving Pre-training by Representing and Predicting Spans","date":"2019-07-24","rows_on_this_dataset":1,"code_links":6,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":15,"samples_ran":6,"samples_unverified":9,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/xlnet-generalized-autoregressive-pretraining","title":"XLNet: Generalized Autoregressive Pretraining for Language Understanding","date":"2019-06-19","rows_on_this_dataset":1,"code_links":27,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":24,"samples_ran":10,"samples_unverified":14,"pointer_only_for_licence":3,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/ernie-enhanced-language-representation-with","title":"ERNIE: Enhanced Language Representation with Informative Entities","date":"2019-05-17","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":3,"samples_ran":3,"samples_unverified":0,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/bert-pre-training-of-deep-bidirectional","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","date":"2018-10-11","rows_on_this_dataset":1,"code_links":534,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":659,"samples_ran":208,"samples_unverified":451,"pointer_only_for_licence":149,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-25T09:33:49+00:00","papers_with_samples":24,"samples_harvested":1032,"samples_ran":430,"samples_unverified":602,"pointer_only_for_licence":235,"papers_with_no_sample_that_ran":2,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}