{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7C4JK5R3IJKWTC4I3ZBHTQTFPD","short_pith_number":"pith:7C4JK5R3","schema_version":"1.0","canonical_sha256":"f8b895763b4255698b88de4279c26578f2ede7bbd75295984fdc1678e2ce908e","source":{"kind":"arxiv","id":"2407.03056","version":2},"attestation_state":"computed","paper":{"title":"Improving Zero-shot Generalization of Learned Prompts via Unsupervised Knowledge Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alberto Baldrati, Andrew D. Bagdanov, Marco Bertini, Marco Mistretta","submitted_at":"2024-07-03T12:24:40Z","abstract_excerpt":"Vision-Language Models (VLMs) demonstrate remarkable zero-shot generalization to unseen tasks, but fall short of the performance of supervised methods in generalizing to downstream tasks with limited data. Prompt learning is emerging as a parameter-efficient method for adapting VLMs, but state-of-the-art approaches require annotated samples. In this paper we propose a novel approach to prompt learning based on unsupervised knowledge distillation from more powerful models. Our approach, which we call Knowledge Distillation Prompt Learning (KDPL), can be integrated into existing prompt learning "},"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":"2407.03056","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-07-03T12:24:40Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3bb638edf005601d89aabf4d5a023f5dd4195f33efa5473a0cd1c8d2d6f57c18","abstract_canon_sha256":"1318d81d8409aa4d9dfc175f64488149a3337814d11e6e2e20b9b3629cc15a40"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:50:13.775988Z","signature_b64":"gox8QUc5rIblvKWNzUjTGDYD8Nh6aNhTRPfgdV/hTyOKtzvJqsl+5BcBuSg86vW7QwOHMpx3JjtYFubaDUVfDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8b895763b4255698b88de4279c26578f2ede7bbd75295984fdc1678e2ce908e","last_reissued_at":"2026-07-05T08:50:13.775493Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:50:13.775493Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Zero-shot Generalization of Learned Prompts via Unsupervised Knowledge Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alberto Baldrati, Andrew D. Bagdanov, Marco Bertini, Marco Mistretta","submitted_at":"2024-07-03T12:24:40Z","abstract_excerpt":"Vision-Language Models (VLMs) demonstrate remarkable zero-shot generalization to unseen tasks, but fall short of the performance of supervised methods in generalizing to downstream tasks with limited data. Prompt learning is emerging as a parameter-efficient method for adapting VLMs, but state-of-the-art approaches require annotated samples. In this paper we propose a novel approach to prompt learning based on unsupervised knowledge distillation from more powerful models. Our approach, which we call Knowledge Distillation Prompt Learning (KDPL), can be integrated into existing prompt learning "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.03056","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/2407.03056/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":"2407.03056","created_at":"2026-07-05T08:50:13.775552+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.03056v2","created_at":"2026-07-05T08:50:13.775552+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.03056","created_at":"2026-07-05T08:50:13.775552+00:00"},{"alias_kind":"pith_short_12","alias_value":"7C4JK5R3IJKW","created_at":"2026-07-05T08:50:13.775552+00:00"},{"alias_kind":"pith_short_16","alias_value":"7C4JK5R3IJKWTC4I","created_at":"2026-07-05T08:50:13.775552+00:00"},{"alias_kind":"pith_short_8","alias_value":"7C4JK5R3","created_at":"2026-07-05T08:50:13.775552+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/7C4JK5R3IJKWTC4I3ZBHTQTFPD","json":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD.json","graph_json":"https://pith.science/api/pith-number/7C4JK5R3IJKWTC4I3ZBHTQTFPD/graph.json","events_json":"https://pith.science/api/pith-number/7C4JK5R3IJKWTC4I3ZBHTQTFPD/events.json","paper":"https://pith.science/paper/7C4JK5R3"},"agent_actions":{"view_html":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD","download_json":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD.json","view_paper":"https://pith.science/paper/7C4JK5R3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.03056&json=true","fetch_graph":"https://pith.science/api/pith-number/7C4JK5R3IJKWTC4I3ZBHTQTFPD/graph.json","fetch_events":"https://pith.science/api/pith-number/7C4JK5R3IJKWTC4I3ZBHTQTFPD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD/action/storage_attestation","attest_author":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD/action/author_attestation","sign_citation":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD/action/citation_signature","submit_replication":"https://pith.science/pith/7C4JK5R3IJKWTC4I3ZBHTQTFPD/action/replication_record"}},"created_at":"2026-07-05T08:50:13.775552+00:00","updated_at":"2026-07-05T08:50:13.775552+00:00"}