{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FGOXEEPY7TENK4IIIZ7JPYS2D6","short_pith_number":"pith:FGOXEEPY","schema_version":"1.0","canonical_sha256":"299d7211f8fcc8d57108467e97e25a1f8e7cfb1d12899239b2c5ec4a3deca385","source":{"kind":"arxiv","id":"2407.08374","version":3},"attestation_state":"computed","paper":{"title":"Enhancing Robustness of Vision-Language Models through Orthogonality Learning and Self-Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dong Zhao, Elisa Ricci, Jinlong Li, Lin Ma, Nicu Sebe, Zequn Jie","submitted_at":"2024-07-11T10:35:53Z","abstract_excerpt":"Efficient fine-tuning of vision-language models (VLMs) like CLIP for specific downstream tasks is gaining significant attention. Previous works primarily focus on prompt learning to adapt the CLIP into a variety of downstream tasks, however, suffering from task overfitting when fine-tuned on a small data set. In this paper, we introduce an orthogonal fine-tuning method for efficiently fine-tuning pretrained weights and enabling enhanced robustness and generalization, while a self-regularization strategy is further exploited to maintain the stability in terms of zero-shot generalization of VLMs"},"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":"2407.08374","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-07-11T10:35:53Z","cross_cats_sorted":[],"title_canon_sha256":"4ef3febf8e191d0d94453bab411883709160618480850bce67a8599406d2b426","abstract_canon_sha256":"8adadc865aa4c8b532fc41d0168c961fb7c2810868b0ff61443d71f739fb9ee1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:21:06.493347Z","signature_b64":"a8BNvG/FJQoI63fVgFWgGG7MpMTXKo8Kewvz6ldrXM5jM8HzjFcligvAKjafwh4YT2WTXdOmJJpDppVxFK47Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"299d7211f8fcc8d57108467e97e25a1f8e7cfb1d12899239b2c5ec4a3deca385","last_reissued_at":"2026-07-05T09:21:06.492840Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:21:06.492840Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Robustness of Vision-Language Models through Orthogonality Learning and Self-Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dong Zhao, Elisa Ricci, Jinlong Li, Lin Ma, Nicu Sebe, Zequn Jie","submitted_at":"2024-07-11T10:35:53Z","abstract_excerpt":"Efficient fine-tuning of vision-language models (VLMs) like CLIP for specific downstream tasks is gaining significant attention. Previous works primarily focus on prompt learning to adapt the CLIP into a variety of downstream tasks, however, suffering from task overfitting when fine-tuned on a small data set. In this paper, we introduce an orthogonal fine-tuning method for efficiently fine-tuning pretrained weights and enabling enhanced robustness and generalization, while a self-regularization strategy is further exploited to maintain the stability in terms of zero-shot generalization of VLMs"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08374","kind":"arxiv","version":3},"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/2407.08374/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":"2407.08374","created_at":"2026-07-05T09:21:06.492895+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.08374v3","created_at":"2026-07-05T09:21:06.492895+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08374","created_at":"2026-07-05T09:21:06.492895+00:00"},{"alias_kind":"pith_short_12","alias_value":"FGOXEEPY7TEN","created_at":"2026-07-05T09:21:06.492895+00:00"},{"alias_kind":"pith_short_16","alias_value":"FGOXEEPY7TENK4II","created_at":"2026-07-05T09:21:06.492895+00:00"},{"alias_kind":"pith_short_8","alias_value":"FGOXEEPY","created_at":"2026-07-05T09:21:06.492895+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/FGOXEEPY7TENK4IIIZ7JPYS2D6","json":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6.json","graph_json":"https://pith.science/api/pith-number/FGOXEEPY7TENK4IIIZ7JPYS2D6/graph.json","events_json":"https://pith.science/api/pith-number/FGOXEEPY7TENK4IIIZ7JPYS2D6/events.json","paper":"https://pith.science/paper/FGOXEEPY"},"agent_actions":{"view_html":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6","download_json":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6.json","view_paper":"https://pith.science/paper/FGOXEEPY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.08374&json=true","fetch_graph":"https://pith.science/api/pith-number/FGOXEEPY7TENK4IIIZ7JPYS2D6/graph.json","fetch_events":"https://pith.science/api/pith-number/FGOXEEPY7TENK4IIIZ7JPYS2D6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6/action/storage_attestation","attest_author":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6/action/author_attestation","sign_citation":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6/action/citation_signature","submit_replication":"https://pith.science/pith/FGOXEEPY7TENK4IIIZ7JPYS2D6/action/replication_record"}},"created_at":"2026-07-05T09:21:06.492895+00:00","updated_at":"2026-07-05T09:21:06.492895+00:00"}