{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PECZGGJK533VJKCYSCIPKCKBYD","short_pith_number":"pith:PECZGGJK","schema_version":"1.0","canonical_sha256":"790593192aeef754a8589090f50941c0ebd633496b27d0874b32a624444051ef","source":{"kind":"arxiv","id":"2303.05828","version":2},"attestation_state":"computed","paper":{"title":"Adapting Contrastive Language-Image Pretrained (CLIP) Models for Out-of-Distribution Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Felix Michels, Markus Kollmann, Nikolas Adaloglou, Tim Kaiser","submitted_at":"2023-03-10T10:02:18Z","abstract_excerpt":"We present a comprehensive experimental study on pretrained feature extractors for visual out-of-distribution (OOD) detection, focusing on adapting contrastive language-image pretrained (CLIP) models. Without fine-tuning on the training data, we are able to establish a positive correlation ($R^2\\geq0.92$) between in-distribution classification and unsupervised OOD detection for CLIP models in $4$ benchmarks. We further propose a new simple and scalable method called \\textit{pseudo-label probing} (PLP) that adapts vision-language models for OOD detection. Given a set of label names of the train"},"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":"2303.05828","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-10T10:02:18Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d2e8897cdfe995a5906b762a66ef348919eec0fce911200d8eb1f3e52cdcb094","abstract_canon_sha256":"e3bfb472b0e52e8f7e663b9313f2cdd2dae7b14f6cbadcc80d00d765cee3198f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:10:46.659556Z","signature_b64":"pvn0J3L+K/TO3KYZs7qcmi++/sXbp4QEhYqUhNw//Qxshze1kqddZvggHwyqb6q7yXqI9Re8Va3S7ktRQ0+sAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"790593192aeef754a8589090f50941c0ebd633496b27d0874b32a624444051ef","last_reissued_at":"2026-07-05T07:10:46.659102Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:10:46.659102Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adapting Contrastive Language-Image Pretrained (CLIP) Models for Out-of-Distribution Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Felix Michels, Markus Kollmann, Nikolas Adaloglou, Tim Kaiser","submitted_at":"2023-03-10T10:02:18Z","abstract_excerpt":"We present a comprehensive experimental study on pretrained feature extractors for visual out-of-distribution (OOD) detection, focusing on adapting contrastive language-image pretrained (CLIP) models. Without fine-tuning on the training data, we are able to establish a positive correlation ($R^2\\geq0.92$) between in-distribution classification and unsupervised OOD detection for CLIP models in $4$ benchmarks. We further propose a new simple and scalable method called \\textit{pseudo-label probing} (PLP) that adapts vision-language models for OOD detection. Given a set of label names of the train"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.05828","kind":"arxiv","version":2},"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/2303.05828/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":"2303.05828","created_at":"2026-07-05T07:10:46.659159+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.05828v2","created_at":"2026-07-05T07:10:46.659159+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.05828","created_at":"2026-07-05T07:10:46.659159+00:00"},{"alias_kind":"pith_short_12","alias_value":"PECZGGJK533V","created_at":"2026-07-05T07:10:46.659159+00:00"},{"alias_kind":"pith_short_16","alias_value":"PECZGGJK533VJKCY","created_at":"2026-07-05T07:10:46.659159+00:00"},{"alias_kind":"pith_short_8","alias_value":"PECZGGJK","created_at":"2026-07-05T07:10:46.659159+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD","json":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD.json","graph_json":"https://pith.science/api/pith-number/PECZGGJK533VJKCYSCIPKCKBYD/graph.json","events_json":"https://pith.science/api/pith-number/PECZGGJK533VJKCYSCIPKCKBYD/events.json","paper":"https://pith.science/paper/PECZGGJK"},"agent_actions":{"view_html":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD","download_json":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD.json","view_paper":"https://pith.science/paper/PECZGGJK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.05828&json=true","fetch_graph":"https://pith.science/api/pith-number/PECZGGJK533VJKCYSCIPKCKBYD/graph.json","fetch_events":"https://pith.science/api/pith-number/PECZGGJK533VJKCYSCIPKCKBYD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD/action/storage_attestation","attest_author":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD/action/author_attestation","sign_citation":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD/action/citation_signature","submit_replication":"https://pith.science/pith/PECZGGJK533VJKCYSCIPKCKBYD/action/replication_record"}},"created_at":"2026-07-05T07:10:46.659159+00:00","updated_at":"2026-07-05T07:10:46.659159+00:00"}