{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:N4QI6TVBTQIWR2QDPSVT5JGYJQ","short_pith_number":"pith:N4QI6TVB","schema_version":"1.0","canonical_sha256":"6f208f4ea19c1168ea037cab3ea4d84c36b6b13406e2e1ee90676758fb9c09fe","source":{"kind":"arxiv","id":"2304.01195","version":1},"attestation_state":"computed","paper":{"title":"Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MM"],"primary_cat":"cs.CV","authors_text":"Aojun Zhou, Bin Zhao, Bowei He, Dong Wang, Peng Gao, Renrui Zhang, Xiangyang Zhu","submitted_at":"2023-04-03T17:58:54Z","abstract_excerpt":"The popularity of Contrastive Language-Image Pre-training (CLIP) has propelled its application to diverse downstream vision tasks. To improve its capacity on downstream tasks, few-shot learning has become a widely-adopted technique. However, existing methods either exhibit limited performance or suffer from excessive learnable parameters. In this paper, we propose APE, an Adaptive Prior rEfinement method for CLIP's pre-trained knowledge, which achieves superior accuracy with high computational efficiency. Via a prior refinement module, we analyze the inter-class disparity in the downstream dat"},"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":"2304.01195","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-04-03T17:58:54Z","cross_cats_sorted":["cs.AI","cs.MM"],"title_canon_sha256":"4436b131aaca09c46c3cd3b76ab50ddafd892500676481f80bcbe335a2e0664e","abstract_canon_sha256":"d2dc26894bc7306536998d00f7295098be21b77ea60f1dc0c2b5e50a104831b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:57:27.653011Z","signature_b64":"IPXGSTFBIoRa/jl0EAbD+CqCa/4BJ8YQwmZwpDY4f+qIwF2jsPwqhFNbGc0XIFxWBHXaHjlBH8tmZBPhVriXDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f208f4ea19c1168ea037cab3ea4d84c36b6b13406e2e1ee90676758fb9c09fe","last_reissued_at":"2026-07-05T05:57:27.652604Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:57:27.652604Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MM"],"primary_cat":"cs.CV","authors_text":"Aojun Zhou, Bin Zhao, Bowei He, Dong Wang, Peng Gao, Renrui Zhang, Xiangyang Zhu","submitted_at":"2023-04-03T17:58:54Z","abstract_excerpt":"The popularity of Contrastive Language-Image Pre-training (CLIP) has propelled its application to diverse downstream vision tasks. To improve its capacity on downstream tasks, few-shot learning has become a widely-adopted technique. However, existing methods either exhibit limited performance or suffer from excessive learnable parameters. In this paper, we propose APE, an Adaptive Prior rEfinement method for CLIP's pre-trained knowledge, which achieves superior accuracy with high computational efficiency. Via a prior refinement module, we analyze the inter-class disparity in the downstream dat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.01195","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/2304.01195/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":"2304.01195","created_at":"2026-07-05T05:57:27.652655+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.01195v1","created_at":"2026-07-05T05:57:27.652655+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.01195","created_at":"2026-07-05T05:57:27.652655+00:00"},{"alias_kind":"pith_short_12","alias_value":"N4QI6TVBTQIW","created_at":"2026-07-05T05:57:27.652655+00:00"},{"alias_kind":"pith_short_16","alias_value":"N4QI6TVBTQIWR2QD","created_at":"2026-07-05T05:57:27.652655+00:00"},{"alias_kind":"pith_short_8","alias_value":"N4QI6TVB","created_at":"2026-07-05T05:57:27.652655+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.15813","citing_title":"Cross-Modal Few-Shot Learning with Second-Order Neural Ordinary Differential Equations","ref_index":54,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ","json":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ.json","graph_json":"https://pith.science/api/pith-number/N4QI6TVBTQIWR2QDPSVT5JGYJQ/graph.json","events_json":"https://pith.science/api/pith-number/N4QI6TVBTQIWR2QDPSVT5JGYJQ/events.json","paper":"https://pith.science/paper/N4QI6TVB"},"agent_actions":{"view_html":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ","download_json":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ.json","view_paper":"https://pith.science/paper/N4QI6TVB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.01195&json=true","fetch_graph":"https://pith.science/api/pith-number/N4QI6TVBTQIWR2QDPSVT5JGYJQ/graph.json","fetch_events":"https://pith.science/api/pith-number/N4QI6TVBTQIWR2QDPSVT5JGYJQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ/action/storage_attestation","attest_author":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ/action/author_attestation","sign_citation":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ/action/citation_signature","submit_replication":"https://pith.science/pith/N4QI6TVBTQIWR2QDPSVT5JGYJQ/action/replication_record"}},"created_at":"2026-07-05T05:57:27.652655+00:00","updated_at":"2026-07-05T05:57:27.652655+00:00"}