{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IQS5CFEPQSTOK47CX3YJ45UIAB","short_pith_number":"pith:IQS5CFEP","schema_version":"1.0","canonical_sha256":"4425d1148f84a6e573e2bef09e76880040cbe84727f692c0005c47dbe1d98406","source":{"kind":"arxiv","id":"2405.14737","version":2},"attestation_state":"computed","paper":{"title":"CLIPScope: Enhancing Zero-Shot OOD Detection with Bayesian Scoring","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Farshad Khorrami, Hao Fu, Naman Patel, Prashanth Krishnamurthy","submitted_at":"2024-05-23T16:03:55Z","abstract_excerpt":"Detection of out-of-distribution (OOD) samples is crucial for safe real-world deployment of machine learning models. Recent advances in vision language foundation models have made them capable of detecting OOD samples without requiring in-distribution (ID) images. However, these zero-shot methods often underperform as they do not adequately consider ID class likelihoods in their detection confidence scoring. Hence, we introduce CLIPScope, a zero-shot OOD detection approach that normalizes the confidence score of a sample by class likelihoods, akin to a Bayesian posterior update. Furthermore, C"},"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":"2405.14737","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-05-23T16:03:55Z","cross_cats_sorted":[],"title_canon_sha256":"a76a513187cd21128e25ee50b59283781c99a5a7a077c5cbf105c04a2209510c","abstract_canon_sha256":"2738ef061d020353bfe8e3751f2edd3f7089cee7472b4a1f2423be256a5f6d04"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:32.085524Z","signature_b64":"xO4/d7Yk5iBmDpILrI5XKJ1/IzaVFr94hXdTgLvJiadUDevo0IUdKsKKH2yrbF+yuROHKcRKYjnFbIoHcmQFAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4425d1148f84a6e573e2bef09e76880040cbe84727f692c0005c47dbe1d98406","last_reissued_at":"2026-07-05T09:33:32.085101Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:32.085101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CLIPScope: Enhancing Zero-Shot OOD Detection with Bayesian Scoring","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Farshad Khorrami, Hao Fu, Naman Patel, Prashanth Krishnamurthy","submitted_at":"2024-05-23T16:03:55Z","abstract_excerpt":"Detection of out-of-distribution (OOD) samples is crucial for safe real-world deployment of machine learning models. Recent advances in vision language foundation models have made them capable of detecting OOD samples without requiring in-distribution (ID) images. However, these zero-shot methods often underperform as they do not adequately consider ID class likelihoods in their detection confidence scoring. Hence, we introduce CLIPScope, a zero-shot OOD detection approach that normalizes the confidence score of a sample by class likelihoods, akin to a Bayesian posterior update. Furthermore, C"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14737","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/2405.14737/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":"2405.14737","created_at":"2026-07-05T09:33:32.085164+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.14737v2","created_at":"2026-07-05T09:33:32.085164+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14737","created_at":"2026-07-05T09:33:32.085164+00:00"},{"alias_kind":"pith_short_12","alias_value":"IQS5CFEPQSTO","created_at":"2026-07-05T09:33:32.085164+00:00"},{"alias_kind":"pith_short_16","alias_value":"IQS5CFEPQSTOK47C","created_at":"2026-07-05T09:33:32.085164+00:00"},{"alias_kind":"pith_short_8","alias_value":"IQS5CFEP","created_at":"2026-07-05T09:33:32.085164+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.09718","citing_title":"BayesAdapter: enhanced uncertainty estimation in CLIP few-shot adaptation","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB","json":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB.json","graph_json":"https://pith.science/api/pith-number/IQS5CFEPQSTOK47CX3YJ45UIAB/graph.json","events_json":"https://pith.science/api/pith-number/IQS5CFEPQSTOK47CX3YJ45UIAB/events.json","paper":"https://pith.science/paper/IQS5CFEP"},"agent_actions":{"view_html":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB","download_json":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB.json","view_paper":"https://pith.science/paper/IQS5CFEP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.14737&json=true","fetch_graph":"https://pith.science/api/pith-number/IQS5CFEPQSTOK47CX3YJ45UIAB/graph.json","fetch_events":"https://pith.science/api/pith-number/IQS5CFEPQSTOK47CX3YJ45UIAB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB/action/storage_attestation","attest_author":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB/action/author_attestation","sign_citation":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB/action/citation_signature","submit_replication":"https://pith.science/pith/IQS5CFEPQSTOK47CX3YJ45UIAB/action/replication_record"}},"created_at":"2026-07-05T09:33:32.085164+00:00","updated_at":"2026-07-05T09:33:32.085164+00:00"}