{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3AD27GJY2UJQD2UE6KIRL6AO6M","short_pith_number":"pith:3AD27GJY","schema_version":"1.0","canonical_sha256":"d807af9938d51301ea84f29115f80ef32e5ff5a2132131eee368756679965949","source":{"kind":"arxiv","id":"2410.20149","version":1},"attestation_state":"computed","paper":{"title":"AdaNeg: Adaptive Negative Proxy Guided OOD Detection with Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Lei Zhang, Yabin Zhang","submitted_at":"2024-10-26T11:20:02Z","abstract_excerpt":"Recent research has shown that pre-trained vision-language models are effective at identifying out-of-distribution (OOD) samples by using negative labels as guidance. However, employing consistent negative labels across different OOD datasets often results in semantic misalignments, as these text labels may not accurately reflect the actual space of OOD images. To overcome this issue, we introduce \\textit{adaptive negative proxies}, which are dynamically generated during testing by exploring actual OOD images, to align more closely with the underlying OOD label space and enhance the efficacy o"},"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":"2410.20149","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-26T11:20:02Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"62799e1890c98739ba27b09b0c3e5245ceeb6a733d9f5e4acd90d261e12cd9d4","abstract_canon_sha256":"b559a4c66afd698a77d5f6d24debc6784740d93e7203b588f8f522a877c3e821"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:26:39.877282Z","signature_b64":"w/ZTvhnomQ6Pe1zCgln6eN4m7Sb+vnRt0J08Dauy5YIy6XLfmiaFHGZB6d/0rvusfFtQTqh8dfqDV7QKDZZqBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d807af9938d51301ea84f29115f80ef32e5ff5a2132131eee368756679965949","last_reissued_at":"2026-07-05T09:26:39.876788Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:26:39.876788Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AdaNeg: Adaptive Negative Proxy Guided OOD Detection with Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Lei Zhang, Yabin Zhang","submitted_at":"2024-10-26T11:20:02Z","abstract_excerpt":"Recent research has shown that pre-trained vision-language models are effective at identifying out-of-distribution (OOD) samples by using negative labels as guidance. However, employing consistent negative labels across different OOD datasets often results in semantic misalignments, as these text labels may not accurately reflect the actual space of OOD images. To overcome this issue, we introduce \\textit{adaptive negative proxies}, which are dynamically generated during testing by exploring actual OOD images, to align more closely with the underlying OOD label space and enhance the efficacy o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.20149","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/2410.20149/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":"2410.20149","created_at":"2026-07-05T09:26:39.876853+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.20149v1","created_at":"2026-07-05T09:26:39.876853+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.20149","created_at":"2026-07-05T09:26:39.876853+00:00"},{"alias_kind":"pith_short_12","alias_value":"3AD27GJY2UJQ","created_at":"2026-07-05T09:26:39.876853+00:00"},{"alias_kind":"pith_short_16","alias_value":"3AD27GJY2UJQD2UE","created_at":"2026-07-05T09:26:39.876853+00:00"},{"alias_kind":"pith_short_8","alias_value":"3AD27GJY","created_at":"2026-07-05T09:26:39.876853+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.02448","citing_title":"Recent Advances in Out-of-Distribution Detection with CLIP-Like Models: A Survey","ref_index":38,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M","json":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M.json","graph_json":"https://pith.science/api/pith-number/3AD27GJY2UJQD2UE6KIRL6AO6M/graph.json","events_json":"https://pith.science/api/pith-number/3AD27GJY2UJQD2UE6KIRL6AO6M/events.json","paper":"https://pith.science/paper/3AD27GJY"},"agent_actions":{"view_html":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M","download_json":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M.json","view_paper":"https://pith.science/paper/3AD27GJY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.20149&json=true","fetch_graph":"https://pith.science/api/pith-number/3AD27GJY2UJQD2UE6KIRL6AO6M/graph.json","fetch_events":"https://pith.science/api/pith-number/3AD27GJY2UJQD2UE6KIRL6AO6M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M/action/storage_attestation","attest_author":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M/action/author_attestation","sign_citation":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M/action/citation_signature","submit_replication":"https://pith.science/pith/3AD27GJY2UJQD2UE6KIRL6AO6M/action/replication_record"}},"created_at":"2026-07-05T09:26:39.876853+00:00","updated_at":"2026-07-05T09:26:39.876853+00:00"}