{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/theoremqa-a-theorem-driven-question-answering","title":"TheoremQA: A Theorem-driven Question Answering dataset","arxiv_id":"2305.12524","date":"2023-05-21","proceeding":null,"authors":["Wenhu Chen","Ming Yin","Max Ku","Pan Lu","Yixin Wan","Xueguang Ma","Jianyu Xu","Xinyi Wang","Tony Xia"],"abstract":"The recent LLMs like GPT-4 and PaLM-2 have made tremendous progress in solving fundamental math problems like GSM8K by achieving over 90% accuracy. However, their capabilities to solve more challenging math problems which require domain-specific knowledge (i.e. theorem) have yet to be investigated. In this paper, we introduce TheoremQA, the first theorem-driven question-answering dataset designed to evaluate AI models' capabilities to apply theorems to solve challenging science problems. TheoremQA is curated by domain experts containing 800 high-quality questions covering 350 theorems (e.g. Taylor's theorem, Lagrange's theorem, Huffman coding, Quantum Theorem, Elasticity Theorem, etc) from Math, Physics, EE&CS, and Finance. We evaluate a wide spectrum of 16 large language and code models with different prompting strategies like Chain-of-Thoughts and Program-of-Thoughts. We found that GPT-4's capabilities to solve these problems are unparalleled, achieving an accuracy of 51% with Program-of-Thoughts Prompting. All the existing open-sourced models are below 15%, barely surpassing the random-guess baseline. Given the diversity and broad coverage of TheoremQA, we believe it can be used as a better benchmark to evaluate LLMs' capabilities to solve challenging science problems. The data and code are released in https://github.com/wenhuchen/TheoremQA.","url_abs":"https://arxiv.org/abs/2305.12524v3","url_pdf":"https://arxiv.org/pdf/2305.12524v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"theoremqa-a-theorem-driven-question-answering","repo_url":"https://github.com/wenhuchen/theoremqa","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"math","task_name":"Math"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-4","method_name":"GPT-4"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[{"slug":"theoremqa","name":"TheoremQA","full_name":"TheoremQA"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"GPT-4 (PoT)","rank_in_archive_order":1,"of":19,"metrics":{"Accuracy":"52.4"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"GPT-4 (CoT)","rank_in_archive_order":2,"of":19,"metrics":{"Accuracy":"43.8"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"GPT-3.5-turbo (PoT)","rank_in_archive_order":3,"of":19,"metrics":{"Accuracy":"35.6"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"PaLM-2-unicorn (CoT)","rank_in_archive_order":6,"of":19,"metrics":{"Accuracy":"31.8"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"GPT-3.5-turbo (CoT)","rank_in_archive_order":7,"of":19,"metrics":{"Accuracy":"30.2"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"Claude-v1 (PoT)","rank_in_archive_order":10,"of":19,"metrics":{"Accuracy":"25.9"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"Claude-v1 (CoT)","rank_in_archive_order":11,"of":19,"metrics":{"Accuracy":"24.9"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"code-davinci-002","rank_in_archive_order":12,"of":19,"metrics":{"Accuracy":"23.9"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"Claude-instant (CoT)","rank_in_archive_order":13,"of":19,"metrics":{"Accuracy":"23.6"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"text-davinci-003","rank_in_archive_order":14,"of":19,"metrics":{"Accuracy":"22.8"},"uses_additional_data":false},{"leaderboard":"/sota/natural-questions-on-theoremqa","task":"Natural Questions","dataset":"TheoremQA","model":"PaLM-2-bison (CoT)","rank_in_archive_order":15,"of":19,"metrics":{"Accuracy":"21.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.12524","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.12524"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/wenhuchen/theoremqa","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/THUDM/VisualGLM-6B","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_draft_wrong":2,"unverified":8},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1},"found_in_text":{"samples":9,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"51f39dd563f707dd","entry":"create_reader_request","repo":"wenhuchen/theoremqa","repo_kind":"official","path":"zero_shot_eval/run_gpt4.py","file_url":"https://github.com/wenhuchen/theoremqa/blob/HEAD/zero_shot_eval/run_gpt4.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"51f39dd563f707dd"}},{"code_sha256_prefix":"826ceb601fb31250","entry":"parse_text","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"web_demo_hf.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/web_demo_hf.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"826ceb601fb31250"}},{"code_sha256_prefix":"b7583b2677118645","entry":"build_prompt","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"cli_demo_hf.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/cli_demo_hf.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b7583b2677118645"}},{"code_sha256_prefix":"a5e88863540a1e1c","entry":"clear_fn","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"web_demo.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/web_demo.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a5e88863540a1e1c"}},{"code_sha256_prefix":"761f8ed7104063cc","entry":"generate_input","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"model/infer_util.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/model/infer_util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"761f8ed7104063cc"}},{"code_sha256_prefix":"775493b10d572a2a","entry":"get_masks_and_position_ids_glm","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"model/chat.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/model/chat.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"775493b10d572a2a"}},{"code_sha256_prefix":"04cb5cd9927c6c03","entry":"is_chinese","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"model/infer_util.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/model/infer_util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"04cb5cd9927c6c03"}},{"code_sha256_prefix":"922434407b7b0339","entry":"predict","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"web_demo_hf.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/web_demo_hf.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"922434407b7b0339"}},{"code_sha256_prefix":"69186ccf8ef0fe53","entry":"process_image","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"model/chat.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/model/chat.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"69186ccf8ef0fe53"}},{"code_sha256_prefix":"dc21db7cf25bfe72","entry":"process_response","repo":"THUDM/VisualGLM-6B","repo_kind":"found_in_text","path":"model/chat.py","file_url":"https://github.com/THUDM/VisualGLM-6B/blob/HEAD/model/chat.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"dc21db7cf25bfe72"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}