{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QYVCKSH77GLKUE6CH7AJSH6XR4","short_pith_number":"pith:QYVCKSH7","schema_version":"1.0","canonical_sha256":"862a2548fff996aa13c23fc0991fd78f265ef99ab58630d9eee6af17a9c875e2","source":{"kind":"arxiv","id":"2302.05738","version":2},"attestation_state":"computed","paper":{"title":"Cross-Modal Fine-Tuning: Align then Refine","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ameet Talwalkar, Corey Staten, Graham Neubig, Junhong Shen, Liam Li, Lucio M. Dery, Mikhail Khodak","submitted_at":"2023-02-11T16:32:28Z","abstract_excerpt":"Fine-tuning large-scale pretrained models has led to tremendous progress in well-studied modalities such as vision and NLP. However, similar gains have not been observed in many other modalities due to a lack of relevant pretrained models. In this work, we propose ORCA, a general cross-modal fine-tuning framework that extends the applicability of a single large-scale pretrained model to diverse modalities. ORCA adapts to a target task via an align-then-refine workflow: given the target input, ORCA first learns an embedding network that aligns the embedded feature distribution with the pretrain"},"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":"2302.05738","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-02-11T16:32:28Z","cross_cats_sorted":[],"title_canon_sha256":"618abb72ff61df02a84417cd316af5cfe4e69c47cc5f89425f8fc470cd406075","abstract_canon_sha256":"241d8bc3edc74aaf7acdb02fc42d3e40805f7d6ccd50378df7e753418460e5e5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:52:26.012608Z","signature_b64":"iQ4Ryo95NcyOdrI1C28mrIgMD4Ofj/BAMhJKic7X8XdaDHufavTp4J85t+7+9egSZ/ysy/CCInIwRvva+z8/DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"862a2548fff996aa13c23fc0991fd78f265ef99ab58630d9eee6af17a9c875e2","last_reissued_at":"2026-07-05T05:52:26.012156Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:52:26.012156Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cross-Modal Fine-Tuning: Align then Refine","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ameet Talwalkar, Corey Staten, Graham Neubig, Junhong Shen, Liam Li, Lucio M. Dery, Mikhail Khodak","submitted_at":"2023-02-11T16:32:28Z","abstract_excerpt":"Fine-tuning large-scale pretrained models has led to tremendous progress in well-studied modalities such as vision and NLP. However, similar gains have not been observed in many other modalities due to a lack of relevant pretrained models. In this work, we propose ORCA, a general cross-modal fine-tuning framework that extends the applicability of a single large-scale pretrained model to diverse modalities. ORCA adapts to a target task via an align-then-refine workflow: given the target input, ORCA first learns an embedding network that aligns the embedded feature distribution with the pretrain"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.05738","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/2302.05738/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":"2302.05738","created_at":"2026-07-05T05:52:26.012214+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.05738v2","created_at":"2026-07-05T05:52:26.012214+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.05738","created_at":"2026-07-05T05:52:26.012214+00:00"},{"alias_kind":"pith_short_12","alias_value":"QYVCKSH77GLK","created_at":"2026-07-05T05:52:26.012214+00:00"},{"alias_kind":"pith_short_16","alias_value":"QYVCKSH77GLKUE6C","created_at":"2026-07-05T05:52:26.012214+00:00"},{"alias_kind":"pith_short_8","alias_value":"QYVCKSH7","created_at":"2026-07-05T05:52:26.012214+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/QYVCKSH77GLKUE6CH7AJSH6XR4","json":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4.json","graph_json":"https://pith.science/api/pith-number/QYVCKSH77GLKUE6CH7AJSH6XR4/graph.json","events_json":"https://pith.science/api/pith-number/QYVCKSH77GLKUE6CH7AJSH6XR4/events.json","paper":"https://pith.science/paper/QYVCKSH7"},"agent_actions":{"view_html":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4","download_json":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4.json","view_paper":"https://pith.science/paper/QYVCKSH7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.05738&json=true","fetch_graph":"https://pith.science/api/pith-number/QYVCKSH77GLKUE6CH7AJSH6XR4/graph.json","fetch_events":"https://pith.science/api/pith-number/QYVCKSH77GLKUE6CH7AJSH6XR4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4/action/storage_attestation","attest_author":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4/action/author_attestation","sign_citation":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4/action/citation_signature","submit_replication":"https://pith.science/pith/QYVCKSH77GLKUE6CH7AJSH6XR4/action/replication_record"}},"created_at":"2026-07-05T05:52:26.012214+00:00","updated_at":"2026-07-05T05:52:26.012214+00:00"}