{"url":"/dataset/scienceqa","name":"ScienceQA","full_name":"Science Question Answering","description_markdown":"**Science Question Answering** (**ScienceQA**) is a new benchmark that consists of 21,208 multimodal multiple choice questions with diverse science topics and annotations of their answers with corresponding lectures and explanations. Out of the questions in **ScienceQA**, 10,332 (48.7%) have an image context, 10,220 (48.2%) have a text context, and 6,532 (30.8%) have both. Most questions are annotated with grounded lectures (83.9%) and detailed explanations (90.5%). The lecture and explanation provide general external knowledge and specific reasons, respectively, for arriving at the correct answer. To the best of our knowledge, **ScienceQA** is the first large-scale multimodal dataset that annotates lectures and explanations for the answers.\r\n\r\n**ScienceQA**, in contrast to previous datasets, has richer domain diversity from three subjects: natural science, language science, and social science. Questions in each subject are categorized first by the topic (Biology, Physics, Chemistry, etc.), then by the category (Plants, Cells, Animals, etc.), and finally by the skill (Classify fruits and vegetables as plant parts, Identify countries of Africa, etc.). **ScienceQA** features 26 topics, 127 categories, and 379 skills that cover a wide range of domains.","description_withheld":null,"homepage":"https://scienceqa.github.io/","introduced_date":"2022-09-20","introduced_date_note":null,"introduced_by":{"paper":"/paper/learn-to-explain-multimodal-reasoning-via","title":"Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering","first_author":"Pan Lu","url":null},"license":{"name":"CC BY-NC-SA","url":null},"modalities":[{"name":"Images","url":"/datasets/modality/images"},{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Question Answering","url":"/task/question-answering","datasets_with_task":"/datasets/task/question-answering"},{"name":"Visual Question Answering (VQA)","url":"/task/visual-question-answering","datasets_with_task":"/datasets/task/visual-question-answering"},{"name":"Multimodal Deep Learning","url":"/task/multimodal-deep-learning","datasets_with_task":"/datasets/task/multimodal-deep-learning"},{"name":"Open-Domain Question Answering","url":"/task/open-domain-question-answering","datasets_with_task":"/datasets/task/open-domain-question-answering"},{"name":"Visual Commonsense Reasoning","url":"/task/visual-commonsense-reasoning","datasets_with_task":"/datasets/task/visual-commonsense-reasoning"},{"name":"Science Question Answering","url":"/task/science-question-answering","datasets_with_task":"/datasets/task/science-question-answering"},{"name":"Explainable Models","url":"/task/explainable-models","datasets_with_task":"/datasets/task/explainable-models"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["ScienceQA"],"data_loaders":[{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/pkulium/D-SQA","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/scillm/ScienceQA","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/derek-thomas/ScienceQA","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/lupantech/ScienceQA","url":"https://drive.google.com/drive/u/1/folders/1w8imCXWYn2LxajmGeGH_g5DaL2rabHev","frameworks":["pytorch"]}],"num_papers_in_archive":339,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/science-question-answering-on-scienceqa","task":"Science Question Answering","dataset_variant":"ScienceQA","rows":10,"metrics":["Avg. Accuracy","Natural Science","Social Science","Language Science","Text Context","Image Context","No Context","Grades 1-6","Grades 7-12"],"first_row_in_archive_order":{"model":"MC-CoT F-Large","paper":"/paper/boosting-the-power-of-small-multimodal","metrics":{"Avg. Accuracy":"94.88","Grades 1-6":"95.3","Grades 7-12":"94.13","Image Context":"93.75","Language Science":"93.18","Natural Science":"97.47","No Context":"94.49","Social Science":"90.44","Text Context":"96.97"},"code_links":[{"title":"chengtan9907/mc-cot","url":"https://github.com/chengtan9907/mc-cot"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/video-lavit-unified-video-language-pre","title":"Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization","date":"2024-02-05","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":3,"samples_unverified":2,"pointer_only_for_licence":5,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/honeybee-locality-enhanced-projector-for","title":"Honeybee: Locality-enhanced Projector for Multimodal LLM","date":"2023-12-11","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":3,"samples_ran":3,"samples_unverified":0,"pointer_only_for_licence":3,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/boosting-the-power-of-small-multimodal","title":"Boosting the Power of Small Multimodal Reasoning Models to Match Larger Models with Self-Consistency Training","date":"2023-11-23","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/chat-univi-unified-visual-representation","title":"Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding","date":"2023-11-14","rows_on_this_dataset":1,"code_links":4,"syntology":null},{"paper":"/paper/multimodal-chain-of-thought-reasoning-in","title":"Multimodal Chain-of-Thought Reasoning in Language Models","date":"2023-02-02","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":11,"samples_ran":6,"samples_unverified":5,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/learn-to-explain-multimodal-reasoning-via","title":"Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering","date":"2022-09-20","rows_on_this_dataset":4,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":4,"samples_unverified":1,"pointer_only_for_licence":2,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":5,"samples_harvested":25,"samples_ran":17,"samples_unverified":8,"pointer_only_for_licence":12,"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."}