{"url":"/task/visual-commonsense-reasoning","name":"Visual Commonsense Reasoning","slug":"visual-commonsense-reasoning","description_markdown":"Image source: [Visual Commonsense Reasoning](https://paperswithcode.com/dataset/vcr)","categories":[{"name":"Computer Vision","url":"/area/computer-vision"},{"name":"Reasoning","url":"/area/reasoning"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":65,"papers_with_code":33,"benchmarks":7,"benchmark_tables_in_archive":7,"benchmark_tables_shown":7,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":8,"subtasks":0,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/visual-commonsense-reasoning-on-gd-vcr","slug":"visual-commonsense-reasoning-on-gd-vcr","dataset":"GD-VCR","dataset_url":"/dataset/gd-vcr","rows_in_archive":4,"metrics":["Accuracy","Gap (West)"],"first_row_in_archive_order":{"model":"Human","paper_title":"Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning","paper_url":"/paper/broaden-the-vision-geo-diverse-visual","paper_date":"2021-09-14","arxiv_id":"2109.06860","code_links":[{"title":"wadeyin9712/gd-vcr","url":"https://github.com/wadeyin9712/gd-vcr"}],"syntology":{"n":9,"n_ran":0,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/visual-commonsense-reasoning-on-vcr-q-a-dev","slug":"visual-commonsense-reasoning-on-vcr-q-a-dev","dataset":"VCR (Q-A) dev","dataset_url":"/dataset/vcr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"PEVL","paper_title":"PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models","paper_url":"/paper/pevl-position-enhanced-pre-training-and","paper_date":"2022-05-23","arxiv_id":"2205.11169","code_links":[{"title":"thunlp/pevl","url":"https://github.com/thunlp/pevl"}],"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":0}}},{"leaderboard":"/sota/visual-commonsense-reasoning-on-vcr-q-a-test","slug":"visual-commonsense-reasoning-on-vcr-q-a-test","dataset":"VCR (Q-A) test","dataset_url":"/dataset/vcr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"PEVL","paper_title":"PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models","paper_url":"/paper/pevl-position-enhanced-pre-training-and","paper_date":"2022-05-23","arxiv_id":"2205.11169","code_links":[{"title":"thunlp/pevl","url":"https://github.com/thunlp/pevl"}],"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":0}}},{"leaderboard":"/sota/visual-commonsense-reasoning-on-vcr-q-ar-dev","slug":"visual-commonsense-reasoning-on-vcr-q-ar-dev","dataset":"VCR (Q-AR) dev","dataset_url":"/dataset/vcr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"PEVL","paper_title":"PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models","paper_url":"/paper/pevl-position-enhanced-pre-training-and","paper_date":"2022-05-23","arxiv_id":"2205.11169","code_links":[{"title":"thunlp/pevl","url":"https://github.com/thunlp/pevl"}],"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":0}}},{"leaderboard":"/sota/visual-commonsense-reasoning-on-vcr-q-ar-test","slug":"visual-commonsense-reasoning-on-vcr-q-ar-test","dataset":"VCR (Q-AR) test","dataset_url":"/dataset/vcr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"PEVL","paper_title":"PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models","paper_url":"/paper/pevl-position-enhanced-pre-training-and","paper_date":"2022-05-23","arxiv_id":"2205.11169","code_links":[{"title":"thunlp/pevl","url":"https://github.com/thunlp/pevl"}],"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":0}}},{"leaderboard":"/sota/visual-commonsense-reasoning-on-vcr-qa-r-dev","slug":"visual-commonsense-reasoning-on-vcr-qa-r-dev","dataset":"VCR (QA-R) dev","dataset_url":"/dataset/vcr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"PEVL","paper_title":"PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models","paper_url":"/paper/pevl-position-enhanced-pre-training-and","paper_date":"2022-05-23","arxiv_id":"2205.11169","code_links":[{"title":"thunlp/pevl","url":"https://github.com/thunlp/pevl"}],"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":0}}},{"leaderboard":"/sota/visual-commonsense-reasoning-on-vcr-qa-r-test","slug":"visual-commonsense-reasoning-on-vcr-qa-r-test","dataset":"VCR (QA-R) test","dataset_url":"/dataset/vcr","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"PEVL","paper_title":"PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models","paper_url":"/paper/pevl-position-enhanced-pre-training-and","paper_date":"2022-05-23","arxiv_id":"2205.11169","code_links":[{"title":"thunlp/pevl","url":"https://github.com/thunlp/pevl"}],"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":0}}}],"datasets":[{"url":"/dataset/scienceqa","name":"ScienceQA","full_name":"Science Question Answering","num_papers_in_archive":339},{"url":"/dataset/vcr","name":"VCR","full_name":"Visual Commonsense Reasoning","num_papers_in_archive":179},{"url":"/dataset/iconqa","name":"IconQA","full_name":"Icon Question Answering","num_papers_in_archive":42},{"url":"/dataset/whoops","name":"WHOOPS!","full_name":"","num_papers_in_archive":14},{"url":"/dataset/gd-vcr","name":"GD-VCR","full_name":"","num_papers_in_archive":6},{"url":"/dataset/polymath","name":"PolyMATH","full_name":"","num_papers_in_archive":1},{"url":"/dataset/qlevr","name":"QLEVR","full_name":"","num_papers_in_archive":1},{"url":"/dataset/visual-commonsense-immorality-benchmark","name":"Visual Commonsense Immorality benchmark","full_name":"","num_papers_in_archive":1}],"subtasks":[],"parent_tasks":[{"url":"/task/visual-reasoning","name":"Visual Reasoning"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":33,"tagged_in_all":65,"items":[{"url":"/paper/vilbert-pretraining-task-agnostic","title":"ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks","date":"2019-08-06","arxiv_id":"1908.02265","repositories_listed":11,"syntology":{"n":34,"n_ran":10,"n_unverified":24,"n_pointer_only":34}},{"url":"/paper/uniter-learning-universal-image-text-1","title":"UNITER: UNiversal Image-TExt Representation Learning","date":"2019-09-25","arxiv_id":"1909.11740","repositories_listed":7,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/making-large-multimodal-models-understand","title":"ViP-LLaVA: Making Large Multimodal Models Understand Arbitrary Visual Prompts","date":"2023-12-01","arxiv_id":"2312.00784","repositories_listed":4,"syntology":null},{"url":"/paper/from-recognition-to-cognition-visual","title":"From Recognition to Cognition: Visual Commonsense Reasoning","date":"2018-11-27","arxiv_id":"1811.10830","repositories_listed":4,"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/gpt4roi-instruction-tuning-large-language","title":"GPT4RoI: Instruction Tuning Large Language Model on Region-of-Interest","date":"2023-07-07","arxiv_id":"2307.03601","repositories_listed":3,"syntology":{"n":5,"n_ran":3,"n_unverified":2,"n_pointer_only":5}},{"url":"/paper/vl-bert-pre-training-of-generic-visual","title":"VL-BERT: Pre-training of Generic Visual-Linguistic Representations","date":"2019-08-22","arxiv_id":"1908.08530","repositories_listed":3,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/x-modaler-a-versatile-and-high-performance","title":"X-modaler: A Versatile and High-performance Codebase for Cross-modal Analytics","date":"2021-08-18","arxiv_id":"2108.08217","repositories_listed":2,"syntology":null},{"url":"/paper/unifying-vision-and-language-tasks-via-text","title":"Unifying Vision-and-Language Tasks via Text Generation","date":"2021-02-04","arxiv_id":"2102.02779","repositories_listed":2,"syntology":{"n":12,"n_ran":2,"n_unverified":10,"n_pointer_only":0}},{"url":"/paper/large-scale-adversarial-training-for-vision","title":"Large-Scale Adversarial Training for Vision-and-Language Representation Learning","date":"2020-06-11","arxiv_id":"2006.06195","repositories_listed":2,"syntology":{"n":20,"n_ran":10,"n_unverified":10,"n_pointer_only":6}},{"url":"/paper/learning-to-correction-explainable-feedback","title":"Learning to Correction: Explainable Feedback Generation for Visual Commonsense Reasoning Distractor","date":"2024-12-08","arxiv_id":"2412.07801","repositories_listed":1,"syntology":null},{"url":"/paper/improving-visual-commonsense-in-language","title":"Improving Visual Commonsense in Language Models via Multiple Image Generation","date":"2024-06-19","arxiv_id":"2406.13621","repositories_listed":1,"syntology":null},{"url":"/paper/dragonfly-multi-resolution-zoom-supercharges","title":"Dragonfly: Multi-Resolution Zoom-In Encoding Enhances Vision-Language Models","date":"2024-06-03","arxiv_id":"2406.00977","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":1}},{"url":"/paper/a-survey-on-interpretable-cross-modal","title":"A Survey on Interpretable Cross-modal Reasoning","date":"2023-09-05","arxiv_id":"2309.01955","repositories_listed":1,"syntology":null},{"url":"/paper/fusing-pre-trained-language-models-with","title":"Fusing Pre-Trained Language Models With Multimodal Prompts Through Reinforcement Learning","date":"2023-01-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/vasr-visual-analogies-of-situation","title":"VASR: Visual Analogies of Situation Recognition","date":"2022-12-08","arxiv_id":"2212.04542","repositories_listed":1,"syntology":{"n":8,"n_ran":7,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/learn-to-explain-multimodal-reasoning-via","title":"Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering","date":"2022-09-20","arxiv_id":"2209.09513","repositories_listed":1,"syntology":{"n":5,"n_ran":4,"n_unverified":1,"n_pointer_only":2}},{"url":"/paper/illume-rationalizing-vision-language-models","title":"ILLUME: Rationalizing Vision-Language Models through Human Interactions","date":"2022-08-17","arxiv_id":"2208.08241","repositories_listed":1,"syntology":null},{"url":"/paper/pevl-position-enhanced-pre-training-and","title":"PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models","date":"2022-05-23","arxiv_id":"2205.11169","repositories_listed":1,"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/vl-interpret-an-interactive-visualization","title":"VL-InterpreT: An Interactive Visualization Tool for Interpreting Vision-Language Transformers","date":"2022-03-30","arxiv_id":"2203.17247","repositories_listed":1,"syntology":{"n":3,"n_ran":2,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/all-in-one-exploring-unified-video-language","title":"All in One: Exploring Unified Video-Language Pre-training","date":"2022-03-14","arxiv_id":"2203.07303","repositories_listed":1,"syntology":null},{"url":"/paper/joint-answering-and-explanation-for-visual","title":"Joint Answering and Explanation for Visual Commonsense Reasoning","date":"2022-02-25","arxiv_id":"2202.12626","repositories_listed":1,"syntology":null},{"url":"/paper/wenlan-2-0-make-ai-imagine-via-a-multimodal","title":"Towards artificial general intelligence via a multimodal foundation model","date":"2021-10-27","arxiv_id":"2110.14378","repositories_listed":1,"syntology":null},{"url":"/paper/broaden-the-vision-geo-diverse-visual","title":"Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning","date":"2021-09-14","arxiv_id":"2109.06860","repositories_listed":1,"syntology":{"n":9,"n_ran":0,"n_unverified":9,"n_pointer_only":0}},{"url":"/paper/interpretable-visual-understanding-with","title":"Interpretable Visual Understanding with Cognitive Attention Network","date":"2021-08-06","arxiv_id":"2108.02924","repositories_listed":1,"syntology":null},{"url":"/paper/cognitive-visual-commonsense-reasoning-using","title":"Cognitive Visual Commonsense Reasoning Using Dynamic Working Memory","date":"2021-07-04","arxiv_id":"2107.01671","repositories_listed":1,"syntology":null},{"url":"/paper/merlot-multimodal-neural-script-knowledge","title":"MERLOT: Multimodal Neural Script Knowledge Models","date":"2021-06-04","arxiv_id":"2106.02636","repositories_listed":1,"syntology":{"n":14,"n_ran":5,"n_unverified":9,"n_pointer_only":0}},{"url":"/paper/natural-language-rationales-with-full-stack","title":"Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs","date":"2020-10-15","arxiv_id":"2010.07526","repositories_listed":1,"syntology":null},{"url":"/paper/connective-cognition-network-for-directional","title":"Connective Cognition Network for Directional Visual Commonsense Reasoning","date":"2019-12-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/tab-vcr-tags-and-attributes-based-vcr-1","title":"TAB-VCR: Tags and Attributes based VCR Baselines","date":"2019-12-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/tab-vcr-tags-and-attributes-based-vcr","title":"TAB-VCR: Tags and Attributes based Visual Commonsense Reasoning Baselines","date":"2019-10-31","arxiv_id":"1910.14671","repositories_listed":1,"syntology":null}],"syntology_records":14,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":1,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}