{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JVBFSJE7CIL7FEK7CGNPXE2PB2","short_pith_number":"pith:JVBFSJE7","schema_version":"1.0","canonical_sha256":"4d4259249f1217f2915f119afb934f0ebf885164571fe2fd655e2c2f52d021a9","source":{"kind":"arxiv","id":"2401.12902","version":1},"attestation_state":"computed","paper":{"title":"Facing the Elephant in the Room: Visual Prompt Tuning or Full Finetuning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cheng Han, Dongfang Liu, Lifu Huang, Qifan Wang, Siyuan Qi, Wenguan Wang, Yiming Cui","submitted_at":"2024-01-23T16:48:18Z","abstract_excerpt":"As the scale of vision models continues to grow, the emergence of Visual Prompt Tuning (VPT) as a parameter-efficient transfer learning technique has gained attention due to its superior performance compared to traditional full-finetuning. However, the conditions favoring VPT (the ``when\") and the underlying rationale (the ``why\") remain unclear. In this paper, we conduct a comprehensive analysis across 19 distinct datasets and tasks. To understand the ``when\" aspect, we identify the scenarios where VPT proves favorable by two dimensions: task objectives and data distributions. We find that VP"},"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":"2401.12902","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-01-23T16:48:18Z","cross_cats_sorted":[],"title_canon_sha256":"1a7884eaa01724b87927e16f68b4e6e71da250c5576f4bc71b8331c550b68d8c","abstract_canon_sha256":"dda8231e561f0a3f72763ad97e8857d140df17a1baf2a4adc29d667a29399e24"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:36:45.410304Z","signature_b64":"LQkwcJgMm9L2HeWJKFQwJQ09yojavj3D28Vr3KsKToVYVlGk3ULG+ITFtCxIwGZsv4BZPxZJE48ohcRrezJ3Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d4259249f1217f2915f119afb934f0ebf885164571fe2fd655e2c2f52d021a9","last_reissued_at":"2026-07-05T07:36:45.409845Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:36:45.409845Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Facing the Elephant in the Room: Visual Prompt Tuning or Full Finetuning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cheng Han, Dongfang Liu, Lifu Huang, Qifan Wang, Siyuan Qi, Wenguan Wang, Yiming Cui","submitted_at":"2024-01-23T16:48:18Z","abstract_excerpt":"As the scale of vision models continues to grow, the emergence of Visual Prompt Tuning (VPT) as a parameter-efficient transfer learning technique has gained attention due to its superior performance compared to traditional full-finetuning. However, the conditions favoring VPT (the ``when\") and the underlying rationale (the ``why\") remain unclear. In this paper, we conduct a comprehensive analysis across 19 distinct datasets and tasks. To understand the ``when\" aspect, we identify the scenarios where VPT proves favorable by two dimensions: task objectives and data distributions. We find that VP"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.12902","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/2401.12902/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":"2401.12902","created_at":"2026-07-05T07:36:45.409909+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.12902v1","created_at":"2026-07-05T07:36:45.409909+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.12902","created_at":"2026-07-05T07:36:45.409909+00:00"},{"alias_kind":"pith_short_12","alias_value":"JVBFSJE7CIL7","created_at":"2026-07-05T07:36:45.409909+00:00"},{"alias_kind":"pith_short_16","alias_value":"JVBFSJE7CIL7FEK7","created_at":"2026-07-05T07:36:45.409909+00:00"},{"alias_kind":"pith_short_8","alias_value":"JVBFSJE7","created_at":"2026-07-05T07:36:45.409909+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11853","citing_title":"Task-Aware Structured Memory for Dynamic Multi-modal In-Context Learning","ref_index":232,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2","json":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2.json","graph_json":"https://pith.science/api/pith-number/JVBFSJE7CIL7FEK7CGNPXE2PB2/graph.json","events_json":"https://pith.science/api/pith-number/JVBFSJE7CIL7FEK7CGNPXE2PB2/events.json","paper":"https://pith.science/paper/JVBFSJE7"},"agent_actions":{"view_html":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2","download_json":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2.json","view_paper":"https://pith.science/paper/JVBFSJE7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.12902&json=true","fetch_graph":"https://pith.science/api/pith-number/JVBFSJE7CIL7FEK7CGNPXE2PB2/graph.json","fetch_events":"https://pith.science/api/pith-number/JVBFSJE7CIL7FEK7CGNPXE2PB2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2/action/storage_attestation","attest_author":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2/action/author_attestation","sign_citation":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2/action/citation_signature","submit_replication":"https://pith.science/pith/JVBFSJE7CIL7FEK7CGNPXE2PB2/action/replication_record"}},"created_at":"2026-07-05T07:36:45.409909+00:00","updated_at":"2026-07-05T07:36:45.409909+00:00"}