{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QCDHBZAGOHSLLX4YKN5ALSTNDO","short_pith_number":"pith:QCDHBZAG","schema_version":"1.0","canonical_sha256":"808670e40671e4b5df98537a05ca6d1b87ec7ff491636544d6b1e5f139253cbd","source":{"kind":"arxiv","id":"2412.10817","version":2},"attestation_state":"computed","paper":{"title":"Enhance Vision-Language Alignment with Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hongyuan Zhang, Sida Huang, Xuelong Li","submitted_at":"2024-12-14T12:58:15Z","abstract_excerpt":"With the advancement of pre-trained vision-language (VL) models, enhancing the alignment between visual and linguistic modalities in downstream tasks has emerged as a critical challenge. Different from existing fine-tuning methods that add extra modules to these two modalities, we investigate whether the frozen model can be fine-tuned by customized noise. Our approach is motivated by the scientific study of beneficial noise, namely Positive-incentive Noise (Pi-noise or $\\pi$-noise) , which quantitatively analyzes the impact of noise. It therefore implies a new scheme to learn beneficial noise "},"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":"2412.10817","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-14T12:58:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9b54799c22bc75d59b0e81008f65d0ac037a20b5dca1ab05dd574f25f1f4633d","abstract_canon_sha256":"d44ac463e60e1ec17b0f9245be4475dc55d84977f54961e947cff3ffaa364d71"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:50:08.906550Z","signature_b64":"wrvgaDf4ds/HHPyU55TtNuCq8LiV2gUmkDGkyWCbrXT1pUN/me4n5yeL2+HBquL1mzGvMPAQm9P07LFEbRqCCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"808670e40671e4b5df98537a05ca6d1b87ec7ff491636544d6b1e5f139253cbd","last_reissued_at":"2026-07-05T09:50:08.905995Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:50:08.905995Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhance Vision-Language Alignment with Noise","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hongyuan Zhang, Sida Huang, Xuelong Li","submitted_at":"2024-12-14T12:58:15Z","abstract_excerpt":"With the advancement of pre-trained vision-language (VL) models, enhancing the alignment between visual and linguistic modalities in downstream tasks has emerged as a critical challenge. Different from existing fine-tuning methods that add extra modules to these two modalities, we investigate whether the frozen model can be fine-tuned by customized noise. Our approach is motivated by the scientific study of beneficial noise, namely Positive-incentive Noise (Pi-noise or $\\pi$-noise) , which quantitatively analyzes the impact of noise. It therefore implies a new scheme to learn beneficial noise "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.10817","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/2412.10817/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":"2412.10817","created_at":"2026-07-05T09:50:08.906056+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.10817v2","created_at":"2026-07-05T09:50:08.906056+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.10817","created_at":"2026-07-05T09:50:08.906056+00:00"},{"alias_kind":"pith_short_12","alias_value":"QCDHBZAGOHSL","created_at":"2026-07-05T09:50:08.906056+00:00"},{"alias_kind":"pith_short_16","alias_value":"QCDHBZAGOHSLLX4Y","created_at":"2026-07-05T09:50:08.906056+00:00"},{"alias_kind":"pith_short_8","alias_value":"QCDHBZAG","created_at":"2026-07-05T09:50:08.906056+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/QCDHBZAGOHSLLX4YKN5ALSTNDO","json":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO.json","graph_json":"https://pith.science/api/pith-number/QCDHBZAGOHSLLX4YKN5ALSTNDO/graph.json","events_json":"https://pith.science/api/pith-number/QCDHBZAGOHSLLX4YKN5ALSTNDO/events.json","paper":"https://pith.science/paper/QCDHBZAG"},"agent_actions":{"view_html":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO","download_json":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO.json","view_paper":"https://pith.science/paper/QCDHBZAG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.10817&json=true","fetch_graph":"https://pith.science/api/pith-number/QCDHBZAGOHSLLX4YKN5ALSTNDO/graph.json","fetch_events":"https://pith.science/api/pith-number/QCDHBZAGOHSLLX4YKN5ALSTNDO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO/action/storage_attestation","attest_author":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO/action/author_attestation","sign_citation":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO/action/citation_signature","submit_replication":"https://pith.science/pith/QCDHBZAGOHSLLX4YKN5ALSTNDO/action/replication_record"}},"created_at":"2026-07-05T09:50:08.906056+00:00","updated_at":"2026-07-05T09:50:08.906056+00:00"}