{"url":"/sota/question-answering-on-trecqa","task":{"name":"Question Answering","url":"/task/question-answering","note":null},"dataset":{"name":"TrecQA","url":"/dataset/trecqa"},"category":"Natural Language Processing","categories":["Miscellaneous","Natural Language Processing","Reasoning"],"category_note":null,"description":"Question answering can be segmented into domain-specific tasks like community question answering and knowledge-base question answering. Popular benchmark datasets for evaluation question answering systems include [SQuAD](/dataset/squad), [HotPotQA](/dataset/hotpotqa), [bAbI](/dataset/babi-1), [TriviaQA](/dataset/triviaqa), [WikiQA](/dataset/wikiqa), and many others. Models for question answering are typically evaluated on metrics like EM and F1. Some recent top performing models are T5 and XLNet.\r\n\r\n( Image credit: [SQuAD](https://rajpurkar.github.io/mlx/qa-and-squad/) )","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["MAP","MRR"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"MAP":"higher","MRR":"higher"}},"counts":{"rows":13,"rows_with_code":7,"rows_with_paper_page":13,"rows_dated":13,"rows_using_additional_data":3},"rows":[{"rank_in_archive_order":1,"model":"TANDA DeBERTa-V3-Large + ALL","metrics":{"MAP":"0.954","MRR":"0.984"},"uses_additional_data":true,"paper_date":"2023-09-15","paper":"/paper/structural-self-supervised-objectives-for","paper_url":"https://arxiv.org/abs/2309.08272v1","paper_title":"Structural Self-Supervised Objectives for Transformers","code":"https://github.com/lucadiliello/transformers-framework","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"TANDA-RoBERTa (ASNQ, TREC-QA)","metrics":{"MAP":"0.943","MRR":"0.974"},"uses_additional_data":true,"paper_date":"2019-11-11","paper":"/paper/tanda-transfer-and-adapt-pre-trained","paper_url":"https://arxiv.org/abs/1911.04118v2","paper_title":"TANDA: Transfer and Adapt Pre-Trained Transformer Models for Answer Sentence Selection","code":"https://github.com/alexa/wqa_tanda","n_code_links":2,"syntology":null},{"rank_in_archive_order":3,"model":"DeBERTa-V3-Large + SSP","metrics":{"MAP":"0.923","MRR":"0.946"},"uses_additional_data":false,"paper_date":"2022-05-20","paper":"/paper/pre-training-transformer-models-with-sentence","paper_url":"https://arxiv.org/abs/2205.10455v2","paper_title":"Pre-training Transformer Models with Sentence-Level Objectives for Answer Sentence Selection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"Contextual DeBERTa-V3-Large + SSP","metrics":{"MAP":"0.919","MRR":"0.945"},"uses_additional_data":true,"paper_date":"2023-05-24","paper":"/paper/context-aware-transformer-pre-training-for","paper_url":"https://arxiv.org/abs/2305.15358v1","paper_title":"Context-Aware Transformer Pre-Training for Answer Sentence Selection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"RLAS-BIABC","metrics":{"MAP":"0.913","MRR":"0.998"},"uses_additional_data":false,"paper_date":"2023-01-07","paper":"/paper/rlas-biabc-a-reinforcement-learning-based","paper_url":"https://arxiv.org/abs/2301.02807v1","paper_title":"RLAS-BIABC: A Reinforcement Learning-Based Answer Selection Using the BERT Model Boosted by an Improved ABC Algorithm","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"RoBERTa-Base Joint + MSPP","metrics":{"MAP":"0.911","MRR":"0.952"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/paragraph-based-transformer-pre-training-for","paper_url":"https://arxiv.org/abs/2205.01228v2","paper_title":"Paragraph-based Transformer Pre-training for Multi-Sentence Inference","code":"https://github.com/amazon-research/wqa-multi-sentence-inference","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"RoBERTa-Base + PSD","metrics":{"MAP":"0.903","MRR":"0.951"},"uses_additional_data":false,"paper_date":"2022-05-20","paper":"/paper/pre-training-transformer-models-with-sentence","paper_url":"https://arxiv.org/abs/2205.10455v2","paper_title":"Pre-training Transformer Models with Sentence-Level Objectives for Answer Sentence Selection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"Comp-Clip + LM + LC","metrics":{"MAP":"0.868","MRR":"0.928"},"uses_additional_data":false,"paper_date":"2019-05-30","paper":"/paper/a-compare-aggregate-model-with-latent","paper_url":"https://arxiv.org/abs/1905.12897v2","paper_title":"A Compare-Aggregate Model with Latent Clustering for Answer Selection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"NLP-Capsule","metrics":{"MAP":"0.7773","MRR":"0.7416"},"uses_additional_data":false,"paper_date":"2019-06-06","paper":"/paper/towards-scalable-and-reliable-capsule","paper_url":"https://arxiv.org/abs/1906.02829v1","paper_title":"Towards Scalable and Reliable Capsule Networks for Challenging NLP Applications","code":"https://github.com/andyweizhao/capsule_text_classification","n_code_links":5,"syntology":null},{"rank_in_archive_order":10,"model":"HyperQA","metrics":{"MAP":"0.770","MRR":"0.825"},"uses_additional_data":false,"paper_date":"2017-07-25","paper":"/paper/hyperbolic-representation-learning-for-fast","paper_url":"http://arxiv.org/abs/1707.07847v3","paper_title":"Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering","code":"https://github.com/vanzytay/WSDM2018_HyperQA","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"PWIN","metrics":{"MAP":"0.7588","MRR":"0.8219"},"uses_additional_data":false,"paper_date":"2016-06-01","paper":"/paper/pairwise-word-interaction-modeling-with-deep","paper_url":"https://aclanthology.org/N16-1108","paper_title":"Pairwise Word Interaction Modeling with Deep Neural Networks for Semantic Similarity Measurement","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"aNMM","metrics":{"MAP":"0.750","MRR":"0.811"},"uses_additional_data":false,"paper_date":"2018-01-05","paper":"/paper/anmm-ranking-short-answer-texts-with","paper_url":"https://arxiv.org/abs/1801.01641v2","paper_title":"aNMM: Ranking Short Answer Texts with Attention-Based Neural Matching Model","code":"https://github.com/yangliuy/aNMM-CIKM16","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"CNN","metrics":{"MAP":"0.711","MRR":"0.785"},"uses_additional_data":false,"paper_date":"2014-12-04","paper":"/paper/deep-learning-for-answer-sentence-selection","paper_url":"http://arxiv.org/abs/1412.1632v1","paper_title":"Deep Learning for Answer Sentence Selection","code":"https://github.com/brmson/dataset-sts","n_code_links":2,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,821 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6821,"papers_extracted_not_yet_verified":65,"boards_without_verdict":29,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":0,"rows_with_any_sample_ran":0,"distinct_papers_with_graph_line":0,"distinct_papers_with_any_sample_ran":0,"samples_over_distinct_papers":{"n_ran":0,"n_unverified":0,"n_samples":0,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":0,"n_unverified":0,"n_samples":0,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}