{"url":"/sota/math-word-problem-solving-on-svamp","task":{"name":"Math Word Problem Solving","url":"/task/math-word-problem-solving","note":null},"dataset":{"name":"SVAMP","url":"/dataset/svamp"},"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":["Execution Accuracy","Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Execution Accuracy":"higher","Accuracy":"higher"}},"counts":{"rows":26,"rows_with_code":25,"rows_with_paper_page":25,"rows_dated":25,"rows_using_additional_data":11},"rows":[{"rank_in_archive_order":1,"model":"GPT-4 (Teaching-Inspired)","metrics":{"Execution Accuracy":"93.9"},"uses_additional_data":false,"paper_date":"2024-10-10","paper":"/paper/teaching-inspired-integrated-prompting","paper_url":"https://arxiv.org/abs/2410.08068v1","paper_title":"Teaching-Inspired Integrated Prompting Framework: A Novel Approach for Enhancing Reasoning in Large Language Models","code":"https://github.com/sallytan13/teaching-inspired-prompting","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":0,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":2,"model":"GPT-4 (Model Selection)","metrics":{"Execution Accuracy":"93.7"},"uses_additional_data":false,"paper_date":"2023-05-23","paper":"/paper/automatic-model-selection-with-large-language","paper_url":"https://arxiv.org/abs/2305.14333v2","paper_title":"Automatic Model Selection with Large Language Models for Reasoning","code":"https://github.com/xuzhao0/model-selection-reasoning","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":3,"model":"Qwen2(CoT + Code Interpreter)","metrics":{"Execution Accuracy":"92.3"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"GPT-4 (PHP)","metrics":{"Execution Accuracy":"91.9"},"uses_additional_data":false,"paper_date":"2023-04-19","paper":"/paper/progressive-hint-prompting-improves-reasoning","paper_url":"https://arxiv.org/abs/2304.09797v6","paper_title":"Progressive-Hint Prompting Improves Reasoning in Large Language Models","code":"https://github.com/chuanyang-Zheng/Progressive-Hint","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":1,"n_samples":4,"n_pointer_only_licence":4}},{"rank_in_archive_order":5,"model":"OpenMath-CodeLlama-70B (w/ code)","metrics":{"Execution Accuracy":"87.8"},"uses_additional_data":true,"paper_date":"2024-02-15","paper":"/paper/openmathinstruct-1-a-1-8-million-math","paper_url":"https://arxiv.org/abs/2402.10176v2","paper_title":"OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset","code":"https://github.com/kipok/nemo-skills","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"MathCoder-L-70B","metrics":{"Execution Accuracy":"84.9"},"uses_additional_data":true,"paper_date":"2023-10-05","paper":"/paper/mathcoder-seamless-code-integration-in-llms","paper_url":"https://arxiv.org/abs/2310.03731v1","paper_title":"MathCoder: Seamless Code Integration in LLMs for Enhanced Mathematical Reasoning","code":"https://github.com/mathllm/mathcoder","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"PoT_Eng (self-consistency @ 5)","metrics":{"Execution Accuracy":"83.70"},"uses_additional_data":false,"paper_date":"2023-06-03","paper":"/paper/does-chatgpt-comprehend-the-place-value-in","paper_url":"https://openreview.net/forum?id=1YiIUJH14U","paper_title":"Does ChatGPT Comprehend the Place Value in Numbers When Solving Math Word Problems?","code":"https://github.com/ajs7270/placevalue-benchmark","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"CoT_Eng (self-consistency @ 5)","metrics":{"Execution Accuracy":"82.50"},"uses_additional_data":false,"paper_date":"2023-06-03","paper":"/paper/does-chatgpt-comprehend-the-place-value-in","paper_url":"https://openreview.net/forum?id=1YiIUJH14U","paper_title":"Does ChatGPT Comprehend the Place Value in Numbers When Solving Math Word Problems?","code":"https://github.com/ajs7270/placevalue-benchmark","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"MMOS-CODE-34B(0-shot)","metrics":{"Execution Accuracy":"80.6"},"uses_additional_data":true,"paper_date":"2024-02-23","paper":"/paper/an-empirical-study-of-data-ability-boundary","paper_url":"https://arxiv.org/abs/2403.00799v1","paper_title":"An Empirical Study of Data Ability Boundary in LLMs' Math Reasoning","code":"https://github.com/cyzhh/MMOS","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":2,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":10,"model":"MMOS-DeepSeekMath-7B(0-shot)","metrics":{"Execution Accuracy":"79.3"},"uses_additional_data":true,"paper_date":"2024-02-23","paper":"/paper/an-empirical-study-of-data-ability-boundary","paper_url":"https://arxiv.org/abs/2403.00799v1","paper_title":"An Empirical Study of Data Ability Boundary in LLMs' Math Reasoning","code":"https://github.com/cyzhh/MMOS","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":2,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":11,"model":"MMOS-CODE-7B(0-shot)","metrics":{"Execution Accuracy":"76.4"},"uses_additional_data":true,"paper_date":"2024-02-23","paper":"/paper/an-empirical-study-of-data-ability-boundary","paper_url":"https://arxiv.org/abs/2403.00799v1","paper_title":"An Empirical Study of Data Ability Boundary in LLMs' Math Reasoning","code":"https://github.com/cyzhh/MMOS","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":2,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":12,"model":"LLaMA 2-Chat","metrics":{"Execution Accuracy":"69.2"},"uses_additional_data":false,"paper_date":"2023-07-18","paper":"/paper/llama-2-open-foundation-and-fine-tuned-chat","paper_url":"https://arxiv.org/abs/2307.09288v2","paper_title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","code":"https://github.com/facebookresearch/llama","n_code_links":19,"syntology":{"n_ran":31,"n_unverified":21,"n_samples":52,"n_pointer_only_licence":16}},{"rank_in_archive_order":13,"model":"DeBERTa","metrics":{"Accuracy":"63.5","Execution Accuracy":"63.5"},"uses_additional_data":false,"paper_date":"2023-06-24","paper":"/paper/math-word-problem-solving-by-generating","paper_url":"https://arxiv.org/abs/2306.13899v1","paper_title":"Math Word Problem Solving by Generating Linguistic Variants of Problem Statements","code":"https://github.com/starscream-11813/variational-mathematical-reasoning","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"PaLM (zero-shot, CoT)","metrics":{"Execution Accuracy":"62.1"},"uses_additional_data":false,"paper_date":"2022-05-24","paper":"/paper/large-language-models-are-zero-shot-reasoners","paper_url":"https://arxiv.org/abs/2205.11916v4","paper_title":"Large Language Models are Zero-Shot Reasoners","code":"https://github.com/kojima-takeshi188/zero_shot_cot","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":1}},{"rank_in_archive_order":15,"model":"PaLM (zero-shot)","metrics":{"Execution Accuracy":"58.8"},"uses_additional_data":false,"paper_date":"2022-05-24","paper":"/paper/large-language-models-are-zero-shot-reasoners","paper_url":"https://arxiv.org/abs/2205.11916v4","paper_title":"Large Language Models are Zero-Shot Reasoners","code":"https://github.com/kojima-takeshi188/zero_shot_cot","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":1}},{"rank_in_archive_order":16,"model":"SYRELM (Vicuna 13B)","metrics":{"Execution Accuracy":"56.65"},"uses_additional_data":true,"paper_date":"2023-12-09","paper":"/paper/frugal-lms-trained-to-invoke-symbolic-solvers","paper_url":"https://arxiv.org/abs/2312.05571v2","paper_title":"Frugal LMs Trained to Invoke Symbolic Solvers Achieve Parameter-Efficient Arithmetic Reasoning","code":"https://github.com/joykirat18/syrelm","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":17,"model":"ATHENA (roberta-large)","metrics":{"Execution Accuracy":"54.8"},"uses_additional_data":false,"paper_date":"2023-11-02","paper":"/paper/athena-mathematical-reasoning-with-thought","paper_url":"https://arxiv.org/abs/2311.01036v1","paper_title":"ATHENA: Mathematical Reasoning with Thought Expansion","code":"https://github.com/the-jb/athena-math","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"MsAT-DeductReasoner","metrics":{"Execution Accuracy":"48.9"},"uses_additional_data":false,"paper_date":"2023-06-02","paper":"/paper/learning-multi-step-reasoning-from-arithmetic","paper_url":"https://arxiv.org/abs/2306.01707v3","paper_title":"Learning Multi-Step Reasoning by Solving Arithmetic Tasks","code":"https://github.com/TianduoWang/MsAT","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"Roberta-DeductReasoner","metrics":{"Execution Accuracy":"47.3"},"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":20,"model":"ATHENA (roberta-base)","metrics":{"Execution Accuracy":"45.6"},"uses_additional_data":false,"paper_date":"2023-11-02","paper":"/paper/athena-mathematical-reasoning-with-thought","paper_url":"https://arxiv.org/abs/2311.01036v1","paper_title":"ATHENA: Mathematical Reasoning with Thought Expansion","code":"https://github.com/the-jb/athena-math","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"Graph2Tree with RoBERTa","metrics":{"Accuracy":"43.8","Execution Accuracy":"43.8"},"uses_additional_data":true,"paper_date":"2021-03-12","paper":"/paper/are-nlp-models-really-able-to-solve-simple","paper_url":"https://arxiv.org/abs/2103.07191v2","paper_title":"Are NLP Models really able to Solve Simple Math Word Problems?","code":"https://github.com/arkilpatel/SVAMP","n_code_links":3,"syntology":null},{"rank_in_archive_order":22,"model":"GTS with RoBERTa","metrics":{"Accuracy":"41.0","Execution Accuracy":"41.0"},"uses_additional_data":true,"paper_date":"2021-03-12","paper":"/paper/are-nlp-models-really-able-to-solve-simple","paper_url":"https://arxiv.org/abs/2103.07191v2","paper_title":"Are NLP Models really able to Solve Simple Math Word Problems?","code":"https://github.com/arkilpatel/SVAMP","n_code_links":3,"syntology":null},{"rank_in_archive_order":23,"model":"LSTM Seq2Seq with RoBERTa","metrics":{"Accuracy":"40.3","Execution Accuracy":"40.3"},"uses_additional_data":true,"paper_date":"2021-03-12","paper":"/paper/are-nlp-models-really-able-to-solve-simple","paper_url":"https://arxiv.org/abs/2103.07191v2","paper_title":"Are NLP Models really able to Solve Simple Math Word Problems?","code":"https://github.com/arkilpatel/SVAMP","n_code_links":3,"syntology":null},{"rank_in_archive_order":24,"model":"SYRELM (GPT-J)","metrics":{"Execution Accuracy":"40.1"},"uses_additional_data":true,"paper_date":"2023-12-09","paper":"/paper/frugal-lms-trained-to-invoke-symbolic-solvers","paper_url":"https://arxiv.org/abs/2312.05571v2","paper_title":"Frugal LMs Trained to Invoke Symbolic Solvers Achieve Parameter-Efficient Arithmetic Reasoning","code":"https://github.com/joykirat18/syrelm","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":25,"model":"Transformer with RoBERTa","metrics":{"Accuracy":"38.9","Execution Accuracy":"38.9"},"uses_additional_data":true,"paper_date":"2021-03-12","paper":"/paper/are-nlp-models-really-able-to-solve-simple","paper_url":"https://arxiv.org/abs/2103.07191v2","paper_title":"Are NLP Models really able to Solve Simple Math Word Problems?","code":"https://github.com/arkilpatel/SVAMP","n_code_links":3,"syntology":null},{"rank_in_archive_order":26,"model":"GPT-4 DUP","metrics":{"Accuracy":"94.2"},"uses_additional_data":false,"paper_date":"2024-04-23","paper":"/paper/achieving-97-on-gsm8k-deeply-understanding","paper_url":"https://arxiv.org/abs/2404.14963v5","paper_title":"Achieving >97% on GSM8K: Deeply Understanding the Problems Makes LLMs Better Solvers for Math Word Problems","code":"https://github.com/whu-zqh/dup","n_code_links":1,"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 6,264 of the 9,581 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":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"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":13,"rows_with_any_sample_ran":11,"distinct_papers_with_graph_line":9,"distinct_papers_with_any_sample_ran":8,"samples_over_distinct_papers":{"n_ran":58,"n_unverified":29,"n_samples":87,"n_pointer_only_licence":46,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":78,"n_unverified":37,"n_samples":115,"n_pointer_only_licence":71,"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"}}}