{"url":"/sota/math-word-problem-solving-on-mathqa","task":{"name":"Math Word Problem Solving","url":"/task/math-word-problem-solving","note":null},"dataset":{"name":"MathQA","url":"/dataset/mathqa"},"category":"Reasoning","categories":["Reasoning"],"category_note":null,"description":"A math word problem is a mathematical exercise (such as in a textbook, worksheet, or exam) where significant background information on the problem is presented in ordinary language rather than in mathematical notation. As most word problems involve a narrative of some sort, they are sometimes referred to as story problems and may vary in the amount of technical language used.","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":["Answer Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Answer Accuracy":"higher"}},"counts":{"rows":5,"rows_with_code":5,"rows_with_paper_page":5,"rows_dated":5,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"ELASTIC (RoBERTa-large)","metrics":{"Answer Accuracy":"83.00"},"uses_additional_data":false,"paper_date":"2022-10-18","paper":"/paper/elastic-numerical-reasoning-with-adaptive","paper_url":"https://arxiv.org/abs/2210.10105v2","paper_title":"ELASTIC: Numerical Reasoning with Adaptive Symbolic Compiler","code":"https://github.com/neurasearch/neurips-2022-submission-3358","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"Exp-Tree","metrics":{"Answer Accuracy":"81.5"},"uses_additional_data":false,"paper_date":"2023-10-14","paper":"/paper/an-expression-tree-decoding-strategy-for","paper_url":"https://arxiv.org/abs/2310.09619v3","paper_title":"An Expression Tree Decoding Strategy for Mathematical Equation Generation","code":"https://github.com/zwq2018/multi-view-consistency-for-mwp","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"Multi-view","metrics":{"Answer Accuracy":"80.6"},"uses_additional_data":true,"paper_date":"2022-10-21","paper":"/paper/multi-view-reasoning-consistent-contrastive","paper_url":"https://arxiv.org/abs/2210.11694v2","paper_title":"Multi-View Reasoning: Consistent Contrastive Learning for Math Word Problem","code":"https://github.com/zwq2018/multi-view-consistency-for-mwp","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":12,"n_samples":13,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"Roberta-DeductReasoner","metrics":{"Answer Accuracy":"78.6"},"uses_additional_data":false,"paper_date":"2022-03-19","paper":"/paper/learning-to-reason-deductively-math-word","paper_url":"https://arxiv.org/abs/2203.10316v4","paper_title":"Learning to Reason Deductively: Math Word Problem Solving as Complex Relation Extraction","code":"https://github.com/allanj/deductive-mwp","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":5,"model":"MWP-BERT","metrics":{"Answer Accuracy":"76.6"},"uses_additional_data":false,"paper_date":"2021-07-28","paper":"/paper/mwp-bert-a-strong-baseline-for-math-word","paper_url":"https://arxiv.org/abs/2107.13435v2","paper_title":"MWP-BERT: Numeracy-Augmented Pre-training for Math Word Problem Solving","code":"https://github.com/lzhenwen/mwp-bert","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":4}}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"syntology":{"read_at":"2026-09-24T18:15:14+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":4,"rows_with_any_sample_ran":4,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":6,"n_unverified":16,"n_samples":22,"n_pointer_only_licence":7,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":6,"n_unverified":16,"n_samples":22,"n_pointer_only_licence":7,"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"}}}