{"url":"/dataset/finqa","name":"FinQA","full_name":null,"description_markdown":"FinQA is a new large-scale dataset with Question-Answering pairs over Financial reports, written by financial experts. The dataset contains 8,281 financial QA\r\npairs, along with their numerical reasoning processes.","description_withheld":null,"homepage":"https://github.com/czyssrs/FinQA","introduced_date":"2021-09-01","introduced_date_note":null,"introduced_by":{"paper":"/paper/finqa-a-dataset-of-numerical-reasoning-over","title":"FinQA: A Dataset of Numerical Reasoning over Financial Data","first_author":"Zhiyu Chen","url":null},"license":{"name":"MIT License","url":"https://github.com/czyssrs/FinQA/blob/main/LICENSE"},"modalities":[{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Question Answering","url":"/task/question-answering","datasets_with_task":"/datasets/task/question-answering"}],"languages":[],"variants":["FinQA"],"data_loaders":[],"num_papers_in_archive":110,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/question-answering-on-finqa","task":"Question Answering","dataset_variant":"FinQA","rows":6,"metrics":["Execution Accuracy","Program Accuracy"],"first_row_in_archive_order":{"model":"APOLLO","paper":"/paper/apollo-an-optimized-training-approach-for","metrics":{"Execution Accuracy":"71.07","Program Accuracy":"68.94"},"code_links":[{"title":"gasolsun36/iter-cot","url":"https://github.com/gasolsun36/iter-cot"},{"title":"gasolsun36/dynamicrag","url":"https://github.com/gasolsun36/dynamicrag"},{"title":"gasolsun36/apollo","url":"https://github.com/gasolsun36/apollo"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/are-chatgpt-and-gpt-4-general-purpose-solvers","title":"Are ChatGPT and GPT-4 General-Purpose Solvers for Financial Text Analytics? A Study on Several Typical Tasks","date":"2023-05-10","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/apollo-an-optimized-training-approach-for","title":"APOLLO: An Optimized Training Approach for Long-form Numerical Reasoning","date":"2022-12-14","rows_on_this_dataset":1,"code_links":3,"syntology":null},{"paper":"/paper/elastic-numerical-reasoning-with-adaptive","title":"ELASTIC: Numerical Reasoning with Adaptive Symbolic Compiler","date":"2022-10-18","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":2,"samples_ran":2,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/finqa-a-dataset-of-numerical-reasoning-over","title":"FinQA: A Dataset of Numerical Reasoning over Financial Data","date":"2021-09-01","rows_on_this_dataset":3,"code_links":1,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":3,"samples_ran":3,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-25T09:33:49+00:00","papers_with_samples":2,"samples_harvested":5,"samples_ran":5,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}