{"url":"/sota/visual-question-answering-on-coco-visual-1","task":{"name":"Visual Question Answering (VQA)","url":"/task/visual-question-answering","note":null},"dataset":{"name":"COCO Visual Question Answering (VQA) real images 1.0 multiple choice","url":"/dataset/coco"},"category":"Computer Vision","categories":["Computer Vision","Natural Language Processing"],"category_note":null,"description":"**Visual Question Answering (VQA)** is a task in computer vision that involves answering questions about an image. The goal of VQA is to teach machines to understand the content of an image and answer questions about it in natural language.\r\n\r\nImage Source: [visualqa.org](https://visualqa.org/)","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":["Percentage correct"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Percentage correct":null}},"counts":{"rows":10,"rows_with_code":8,"rows_with_paper_page":10,"rows_dated":10,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"MCB 7 att.","metrics":{"Percentage correct":"70.1"},"uses_additional_data":false,"paper_date":"2016-06-06","paper":"/paper/multimodal-compact-bilinear-pooling-for","paper_url":"http://arxiv.org/abs/1606.01847v3","paper_title":"Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding","code":"https://github.com/Cadene/vqa.pytorch","n_code_links":10,"syntology":null},{"rank_in_archive_order":2,"model":"Dual-MFA","metrics":{"Percentage correct":"70.04"},"uses_additional_data":false,"paper_date":"2017-11-18","paper":"/paper/co-attending-free-form-regions-and-detections","paper_url":"http://arxiv.org/abs/1711.06794v2","paper_title":"Co-attending Free-form Regions and Detections with Multi-modal Multiplicative Feature Embedding for Visual Question Answering","code":"https://github.com/lupantech/dual-mfa-vqa","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"RelAtt","metrics":{"Percentage correct":"69.60"},"uses_additional_data":false,"paper_date":"2018-05-24","paper":"/paper/r-vqa-learning-visual-relation-facts-with","paper_url":"http://arxiv.org/abs/1805.09701v2","paper_title":"R-VQA: Learning Visual Relation Facts with Semantic Attention for Visual Question Answering","code":"https://github.com/lupantech/rvqa","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"3-Modalities: Unary + Pairwise + Ternary (ResNet)","metrics":{"Percentage correct":"69.3"},"uses_additional_data":false,"paper_date":"2017-11-12","paper":"/paper/high-order-attention-models-for-visual","paper_url":"http://arxiv.org/abs/1711.04323v1","paper_title":"High-Order Attention Models for Visual Question Answering","code":"https://github.com/idansc/HighOrderAtten","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"joint-loss","metrics":{"Percentage correct":"67.3"},"uses_additional_data":false,"paper_date":"2016-06-12","paper":"/paper/training-recurrent-answering-units-with-joint","paper_url":"http://arxiv.org/abs/1606.03647v2","paper_title":"Training Recurrent Answering Units with Joint Loss Minimization for VQA","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"MRN","metrics":{"Percentage correct":"66.3"},"uses_additional_data":false,"paper_date":"2016-06-05","paper":"/paper/multimodal-residual-learning-for-visual-qa","paper_url":"http://arxiv.org/abs/1606.01455v2","paper_title":"Multimodal Residual Learning for Visual QA","code":"https://github.com/jnhwkim/nips-mrn-vqa","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"HQI+ResNet","metrics":{"Percentage correct":"66.1"},"uses_additional_data":false,"paper_date":"2016-05-31","paper":"/paper/hierarchical-question-image-co-attention-for","paper_url":"http://arxiv.org/abs/1606.00061v5","paper_title":"Hierarchical Question-Image Co-Attention for Visual Question Answering","code":"https://github.com/jiasenlu/HieCoAttenVQA","n_code_links":9,"syntology":{"n_ran":1,"n_unverified":6,"n_samples":7,"n_pointer_only_licence":1}},{"rank_in_archive_order":8,"model":"FDA","metrics":{"Percentage correct":"64.2"},"uses_additional_data":false,"paper_date":"2016-04-06","paper":"/paper/a-focused-dynamic-attention-model-for-visual","paper_url":"http://arxiv.org/abs/1604.01485v1","paper_title":"A Focused Dynamic Attention Model for Visual Question Answering","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"LSTM Q+I","metrics":{"Percentage correct":"63.1"},"uses_additional_data":false,"paper_date":"2015-05-03","paper":"/paper/vqa-visual-question-answering","paper_url":"http://arxiv.org/abs/1505.00468v7","paper_title":"VQA: Visual Question Answering","code":"https://github.com/ramprs/grad-cam","n_code_links":21,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":6}},{"rank_in_archive_order":10,"model":"iBOWIMG baseline","metrics":{"Percentage correct":"62.0"},"uses_additional_data":false,"paper_date":"2015-12-07","paper":"/paper/simple-baseline-for-visual-question-answering","paper_url":"http://arxiv.org/abs/1512.02167v2","paper_title":"Simple Baseline for Visual Question Answering","code":"https://github.com/metalbubble/VQAbaseline","n_code_links":7,"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":2,"rows_with_any_sample_ran":2,"distinct_papers_with_graph_line":2,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":7,"n_unverified":7,"n_samples":14,"n_pointer_only_licence":7,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":7,"n_unverified":7,"n_samples":14,"n_pointer_only_licence":7,"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"}}}