{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/from-recognition-to-cognition-visual","title":"From Recognition to Cognition: Visual Commonsense Reasoning","arxiv_id":"1811.10830","date":"2018-11-27","proceeding":"CVPR 2019 6","authors":["Rowan Zellers","Yonatan Bisk","Ali Farhadi","Yejin Choi"],"abstract":"Visual understanding goes well beyond object recognition. With one glance at\nan image, we can effortlessly imagine the world beyond the pixels: for\ninstance, we can infer people's actions, goals, and mental states. While this\ntask is easy for humans, it is tremendously difficult for today's vision\nsystems, requiring higher-order cognition and commonsense reasoning about the\nworld. We formalize this task as Visual Commonsense Reasoning. Given a\nchallenging question about an image, a machine must answer correctly and then\nprovide a rationale justifying its answer.\n  Next, we introduce a new dataset, VCR, consisting of 290k multiple choice QA\nproblems derived from 110k movie scenes. The key recipe for generating\nnon-trivial and high-quality problems at scale is Adversarial Matching, a new\napproach to transform rich annotations into multiple choice questions with\nminimal bias. Experimental results show that while humans find VCR easy (over\n90% accuracy), state-of-the-art vision models struggle (~45%).\n  To move towards cognition-level understanding, we present a new reasoning\nengine, Recognition to Cognition Networks (R2C), that models the necessary\nlayered inferences for grounding, contextualization, and reasoning. R2C helps\nnarrow the gap between humans and machines (~65%); still, the challenge is far\nfrom solved, and we provide analysis that suggests avenues for future work.","url_abs":"http://arxiv.org/abs/1811.10830v2","url_pdf":"http://arxiv.org/pdf/1811.10830v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"from-recognition-to-cognition-visual","repo_url":"https://github.com/KiritoSSR/paddle_r2c","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"paddle","reach":{"status":"ok"}},{"paper_slug":"from-recognition-to-cognition-visual","repo_url":"https://github.com/TheShadow29/visual-commonsense-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"from-recognition-to-cognition-visual","repo_url":"https://github.com/karanchahal/play-with-vcr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"from-recognition-to-cognition-visual","repo_url":"https://github.com/rowanz/r2c","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"multiple-choice-qa","task_name":"Multiple Choice Question Answering (MCQA)"},{"task_slug":"multiple-choice","task_name":"Multiple-choice"},{"task_slug":"visual-commonsense-reasoning","task_name":"Visual Commonsense Reasoning"}],"methods":[],"datasets_introduced":[{"slug":"vcr","name":"VCR","full_name":"Visual Commonsense Reasoning"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.10830","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1811.10830"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/TheShadow29/visual-commonsense-pytorch","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/karanchahal/play-with-vcr","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/KiritoSSR/paddle_r2c","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/rowanz/r2c","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":6},"by_repo_kind":{"listed":{"samples":6,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"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"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"554b63fcc01c5a02","entry":"clip_grad_norm","repo":"karanchahal/play-with-vcr","repo_kind":"listed","path":"utils/pytorch_misc.py","file_url":"https://github.com/karanchahal/play-with-vcr/blob/HEAD/utils/pytorch_misc.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"554b63fcc01c5a02"}},{"code_sha256_prefix":"6b16563fec0d7893","entry":"extra_leading_dim_in_sequence","repo":"karanchahal/play-with-vcr","repo_kind":"listed","path":"utils/pytorch_misc.py","file_url":"https://github.com/karanchahal/play-with-vcr/blob/HEAD/utils/pytorch_misc.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6b16563fec0d7893"}},{"code_sha256_prefix":"3781075867d2d4ea","entry":"load_image","repo":"karanchahal/play-with-vcr","repo_kind":"listed","path":"dataloaders/box_utils.py","file_url":"https://github.com/karanchahal/play-with-vcr/blob/HEAD/dataloaders/box_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3781075867d2d4ea"}},{"code_sha256_prefix":"baaa4cdec3d1c236","entry":"make_mask","repo":"karanchahal/play-with-vcr","repo_kind":"listed","path":"dataloaders/mask_utils.py","file_url":"https://github.com/karanchahal/play-with-vcr/blob/HEAD/dataloaders/mask_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"baaa4cdec3d1c236"}},{"code_sha256_prefix":"154ac482998c4966","entry":"pad_sequence","repo":"karanchahal/play-with-vcr","repo_kind":"listed","path":"utils/pytorch_misc.py","file_url":"https://github.com/karanchahal/play-with-vcr/blob/HEAD/utils/pytorch_misc.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"154ac482998c4966"}},{"code_sha256_prefix":"086cec07d5d5e16d","entry":"resize_image","repo":"karanchahal/play-with-vcr","repo_kind":"listed","path":"dataloaders/box_utils.py","file_url":"https://github.com/karanchahal/play-with-vcr/blob/HEAD/dataloaders/box_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"086cec07d5d5e16d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}