Papers › TheoremQA: A Theorem-driven Question Answering dataset

TheoremQA: A Theorem-driven Question Answering dataset

21 May 2023arXiv:2305.12524archive 2025-07-28

Wenhu Chen, Ming Yin, Max Ku, Pan Lu, Yixin Wan, Xueguang Ma, Jianyu Xu, Xinyi Wang, Tony Xia

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.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 10 code samples harvested from 2 repositories linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 1 sample from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed); found in paper text by Syntology: 9 samples from 1 repository, 9 ran (7 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

wenhuchen/theoremqa officialmentioned in papermentioned on GitHubpytorchSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
7ran

Licence: 0 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: 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. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

wenhuchen/theoremqa official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
create_reader_request zero_shot_eval/run_gpt4.py ran · our draft was wrong MIT (permissive) · 51f39dd563f707dd · report fixture: signature example:dict/0/? · returned str · level recorded 2026-09-22
THUDM/VisualGLM-6B found in paper text by Syntology 9 samples, 9 ran (2 instrument-failed) · 0 unverified
build_prompt in 78 papers cli_demo_hf.py ran Apache-2.0 (permissive) · b7583b2677118645 · report fixture: signature history:list/1/tuple, prefix:str/0/str · returned str · level recorded 2026-09-26
clear_fn web_demo.py ran fingerprinted Apache-2.0 (permissive) · a5e88863540a1e1c · report fixture: signature value:str/0/str · returned tuple · level recorded 2026-09-26
generate_input in 6 papers model/infer_util.py ran Apache-2.0 (permissive) · 761f8ed7104063cc · report fixture: signature input_text:str/0/str, input_image_prompt:str/0/str, history:list/1/dict, input_para:dict/0/?, image_is_encoded:bool/0/bool · returned dict · level recorded 2026-09-26
get_masks_and_position_ids_glm model/chat.py ran Apache-2.0 (permissive) · 775493b10d572a2a · report fixture: signature seq:Tensor/1/int, mask_position:int/0/int, context_length:int/0/int · returned tuple · level recorded 2026-09-26
is_chinese in 6 papers model/infer_util.py ran Apache-2.0 (permissive) · 04cb5cd9927c6c03 · report fixture: signature text:str/0/str · returned Match · level recorded 2026-09-26
parse_text in 15 papers web_demo_hf.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 826ceb601fb31250 · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-22
predict in 126 papers web_demo_hf.py ran Apache-2.0 (permissive) · 922434407b7b0339 · report fixture: signature input:str/0/str, image_path:str/0/str, chatbot:list/1/empty, max_length:int/0/int, top_p:float/0/float, temperature:float/0/float, history:list/1/empty · returned generator · level recorded 2026-09-26
process_image in 43 papers model/chat.py ran Apache-2.0 (permissive) · 69186ccf8ef0fe53 · report fixture: signature text:str/0/str, image:NoneType/0/? · returned tuple · level recorded 2026-09-26
process_response in 18 papers model/chat.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · dc21db7cf25bfe72 · report fixture: signature response:str/0/str · returned str · level recorded 2026-09-25

Tasks

MathQuestion Answering

Datasets

Introduced by this paper, per the archive.

TheoremQA

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Natural Questions TheoremQA GPT-4 (PoT) Accuracy 52.4 #1 of 19 Archive leaderboard report
Natural Questions TheoremQA GPT-4 (CoT) Accuracy 43.8 #2 of 19 Archive leaderboard report
Natural Questions TheoremQA GPT-3.5-turbo (PoT) Accuracy 35.6 #3 of 19 Archive leaderboard report
Natural Questions TheoremQA PaLM-2-unicorn (CoT) Accuracy 31.8 #6 of 19 Archive leaderboard report
Natural Questions TheoremQA GPT-3.5-turbo (CoT) Accuracy 30.2 #7 of 19 Archive leaderboard report
Natural Questions TheoremQA Claude-v1 (PoT) Accuracy 25.9 #10 of 19 Archive leaderboard report
Natural Questions TheoremQA Claude-v1 (CoT) Accuracy 24.9 #11 of 19 Archive leaderboard report
Natural Questions TheoremQA code-davinci-002 Accuracy 23.9 #12 of 19 Archive leaderboard report
Natural Questions TheoremQA Claude-instant (CoT) Accuracy 23.6 #13 of 19 Archive leaderboard report
Natural Questions TheoremQA text-davinci-003 Accuracy 22.8 #14 of 19 Archive leaderboard report
Natural Questions TheoremQA PaLM-2-bison (CoT) Accuracy 21.0 #15 of 19 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGPT-4Label SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections