{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:WDLASH5TNLN5LGIJNQRPTB2AR6","short_pith_number":"pith:WDLASH5T","schema_version":"1.0","canonical_sha256":"b0d6091fb36adbd599096c22f987408f93dec0caa92f3b750a93e990602a8422","source":{"kind":"arxiv","id":"2007.05566","version":1},"attestation_state":"computed","paper":{"title":"Contrastive Training for Improved Out-of-Distribution Detection","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhijit Guha Roy, Alan Karthikesalingam, Jim Winkens, Joseph R. Ledsam, Olaf Ronneberger, Patricia MacWilliams, Pushmeet Kohli, Robert Stanforth, Rudy Bunel, Simon Kohl, S. M. Ali Eslami, Taylan Cemgil, Vivek Natarajan","submitted_at":"2020-07-10T18:40:37Z","abstract_excerpt":"Reliable detection of out-of-distribution (OOD) inputs is increasingly understood to be a precondition for deployment of machine learning systems. This paper proposes and investigates the use of contrastive training to boost OOD detection performance. Unlike leading methods for OOD detection, our approach does not require access to examples labeled explicitly as OOD, which can be difficult to collect in practice. We show in extensive experiments that contrastive training significantly helps OOD detection performance on a number of common benchmarks. By introducing and employing the Confusion L"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2007.05566","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2020-07-10T18:40:37Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"f91f8c80ea790e078dd7d774c003595b756a7273e24369f4aec2d7304d905ede","abstract_canon_sha256":"85b53a9748c5dfffd73a812d1446de3c9a3d2ce6b032afcc03b97f43500e69c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:17:49.148245Z","signature_b64":"iGGcUXT9Jb4iMmGlkvaDs66xe6+04InaSdGvdWLb/9p+wJjDCp0a+QpPFADG94vKPvQcnGqHcQr7x8osAWcjCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0d6091fb36adbd599096c22f987408f93dec0caa92f3b750a93e990602a8422","last_reissued_at":"2026-07-05T01:17:49.147844Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:17:49.147844Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrastive Training for Improved Out-of-Distribution Detection","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhijit Guha Roy, Alan Karthikesalingam, Jim Winkens, Joseph R. Ledsam, Olaf Ronneberger, Patricia MacWilliams, Pushmeet Kohli, Robert Stanforth, Rudy Bunel, Simon Kohl, S. M. Ali Eslami, Taylan Cemgil, Vivek Natarajan","submitted_at":"2020-07-10T18:40:37Z","abstract_excerpt":"Reliable detection of out-of-distribution (OOD) inputs is increasingly understood to be a precondition for deployment of machine learning systems. This paper proposes and investigates the use of contrastive training to boost OOD detection performance. Unlike leading methods for OOD detection, our approach does not require access to examples labeled explicitly as OOD, which can be difficult to collect in practice. We show in extensive experiments that contrastive training significantly helps OOD detection performance on a number of common benchmarks. By introducing and employing the Confusion L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.05566","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2007.05566/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2007.05566","created_at":"2026-07-05T01:17:49.147901+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.05566v1","created_at":"2026-07-05T01:17:49.147901+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.05566","created_at":"2026-07-05T01:17:49.147901+00:00"},{"alias_kind":"pith_short_12","alias_value":"WDLASH5TNLN5","created_at":"2026-07-05T01:17:49.147901+00:00"},{"alias_kind":"pith_short_16","alias_value":"WDLASH5TNLN5LGIJ","created_at":"2026-07-05T01:17:49.147901+00:00"},{"alias_kind":"pith_short_8","alias_value":"WDLASH5T","created_at":"2026-07-05T01:17:49.147901+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24404","citing_title":"Modality-Aware Out-of-Distribution Detection for Multi-Modal Action Recognition","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31155","citing_title":"Learning Hyperspherical Time-Frequency Representations for Time-Series Out-of-Distribution Detection","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27411","citing_title":"Detecting is Easy, Adapting is Hard: Local Expert Growth for Visual Model-Based Reinforcement Learning under Distribution Shift","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06382","citing_title":"Rethinking Vacuity for OOD Detection in Evidential Deep Learning","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2207.05221","citing_title":"Language Models (Mostly) Know What They Know","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6","json":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6.json","graph_json":"https://pith.science/api/pith-number/WDLASH5TNLN5LGIJNQRPTB2AR6/graph.json","events_json":"https://pith.science/api/pith-number/WDLASH5TNLN5LGIJNQRPTB2AR6/events.json","paper":"https://pith.science/paper/WDLASH5T"},"agent_actions":{"view_html":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6","download_json":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6.json","view_paper":"https://pith.science/paper/WDLASH5T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.05566&json=true","fetch_graph":"https://pith.science/api/pith-number/WDLASH5TNLN5LGIJNQRPTB2AR6/graph.json","fetch_events":"https://pith.science/api/pith-number/WDLASH5TNLN5LGIJNQRPTB2AR6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6/action/storage_attestation","attest_author":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6/action/author_attestation","sign_citation":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6/action/citation_signature","submit_replication":"https://pith.science/pith/WDLASH5TNLN5LGIJNQRPTB2AR6/action/replication_record"}},"created_at":"2026-07-05T01:17:49.147901+00:00","updated_at":"2026-07-05T01:17:49.147901+00:00"}