{"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/delving-into-out-of-distribution-detection","title":"Delving into Out-of-Distribution Detection with Vision-Language Representations","arxiv_id":"2211.13445","date":"2022-11-24","proceeding":null,"authors":["Yifei Ming","Ziyang Cai","Jiuxiang Gu","Yiyou Sun","Wei Li","Yixuan Li"],"abstract":"Recognizing out-of-distribution (OOD) samples is critical for machine learning systems deployed in the open world. The vast majority of OOD detection methods are driven by a single modality (e.g., either vision or language), leaving the rich information in multi-modal representations untapped. Inspired by the recent success of vision-language pre-training, this paper enriches the landscape of OOD detection from a single-modal to a multi-modal regime. Particularly, we propose Maximum Concept Matching (MCM), a simple yet effective zero-shot OOD detection method based on aligning visual features with textual concepts. We contribute in-depth analysis and theoretical insights to understand the effectiveness of MCM. Extensive experiments demonstrate that MCM achieves superior performance on a wide variety of real-world tasks. MCM with vision-language features outperforms a common baseline with pure visual features on a hard OOD task with semantically similar classes by 13.1% (AUROC). Code is available at https://github.com/deeplearning-wisc/MCM.","url_abs":"https://arxiv.org/abs/2211.13445v1","url_pdf":"https://arxiv.org/pdf/2211.13445v1.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":"delving-into-out-of-distribution-detection","repo_url":"https://github.com/deeplearning-wisc/mcm","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"delving-into-out-of-distribution-detection","repo_url":"https://github.com/HHU-MMBS/plp-official-tmlr2024","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"out-of-distribution-detection","task_name":"Out-of-Distribution Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-12","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs Curated OODs (avg.)","model":"MCM (CLIP-L)","rank_in_archive_order":10,"of":16,"metrics":{"AUROC":"91.49","FPR95":"38.17"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-9","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs Places","model":"MCM (CLIP-L)","rank_in_archive_order":6,"of":25,"metrics":{"AUROC":"92.00","FPR95":"35.42"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-9","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs Places","model":"MCM (CLIP-B)","rank_in_archive_order":10,"of":25,"metrics":{"AUROC":"89.77","FPR95":"44.69"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-8","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs SUN","model":"MCM (CLIP-L)","rank_in_archive_order":7,"of":22,"metrics":{"AUROC":"94.14","FPR95":"29.00"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-8","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs SUN","model":"MCM (CLIP-B)","rank_in_archive_order":10,"of":22,"metrics":{"AUROC":"92.57","FPR95":"37.59"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-10","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs Textures","model":"MCM (CLIP-B)","rank_in_archive_order":23,"of":34,"metrics":{"AUROC":"86.11","FPR95":"57.77"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-10","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs Textures","model":"MCM (CLIP-L)","rank_in_archive_order":25,"of":34,"metrics":{"AUROC":"84.88","FPR95":"59.88"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-3","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs iNaturalist","model":"MCM (CLIP-L)","rank_in_archive_order":15,"of":28,"metrics":{"AUROC":"94.95","FPR95":"28.38"},"uses_additional_data":false},{"leaderboard":"/sota/out-of-distribution-detection-on-imagenet-1k-3","task":"Out-of-Distribution Detection","dataset":"ImageNet-1k vs iNaturalist","model":"MCM (CLIP-B)","rank_in_archive_order":18,"of":28,"metrics":{"AUROC":"94.61","FPR95":"30.91"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2211.13445","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2211.13445"}},"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":"deterministic:regex_extraction","url":"https://github.com/openai/CLIP","reach":{"status":"unanswered"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/deeplearning-wisc/MCM","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HHU-MMBS/plp-official-tmlr2024","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/deeplearning-wisc/mcm","reach":{"status":"ok"}}],"summary":{"ran_fixture":3,"ran":1,"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1},"listed":{"samples":4,"ran":3,"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":1,"samples":[{"code_sha256_prefix":"b2b3162dc355af43","entry":"OOD_cls_max_val","repo":"HHU-MMBS/plp-official-tmlr2024","repo_kind":"listed","path":"eval/ood_scores.py","file_url":"https://github.com/HHU-MMBS/plp-official-tmlr2024/blob/HEAD/eval/ood_scores.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b2b3162dc355af43"}},{"code_sha256_prefix":"d437442e8b6bc274","entry":"_norm_by_name","repo":"HHU-MMBS/plp-official-tmlr2024","repo_kind":"listed","path":"eval/ood_scores.py","file_url":"https://github.com/HHU-MMBS/plp-official-tmlr2024/blob/HEAD/eval/ood_scores.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d437442e8b6bc274"}},{"code_sha256_prefix":"ba7d44d0dbfa802c","entry":"available_norms","repo":"HHU-MMBS/plp-official-tmlr2024","repo_kind":"listed","path":"eval/ood_scores.py","file_url":"https://github.com/HHU-MMBS/plp-official-tmlr2024/blob/HEAD/eval/ood_scores.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ba7d44d0dbfa802c"}},{"code_sha256_prefix":"44d9ad0d40e49f34","entry":"get_ood_scores_clip","repo":"deeplearning-wisc/mcm","repo_kind":"official","path":"utils/detection_util.py","file_url":"https://github.com/deeplearning-wisc/mcm/blob/HEAD/utils/detection_util.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"44d9ad0d40e49f34"}},{"code_sha256_prefix":"8592992c17099ed5","entry":"norm_feats","repo":"HHU-MMBS/plp-official-tmlr2024","repo_kind":"listed","path":"eval/ood_scores.py","file_url":"https://github.com/HHU-MMBS/plp-official-tmlr2024/blob/HEAD/eval/ood_scores.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8592992c17099ed5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}