{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:X57NCA34DI53BMKEPCHUHAQZQV","short_pith_number":"pith:X57NCA34","schema_version":"1.0","canonical_sha256":"bf7ed1037c1a3bb0b144788f438219855a9f537a52a3b1a6dc6857221941518a","source":{"kind":"arxiv","id":"2412.15652","version":1},"attestation_state":"computed","paper":{"title":"Error-driven Data-efficient Large Multimodal Model Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Barry Menglong Yao (UC Davis), Lifu Huang (UC Davis), Qifan Wang (Meta AI)","submitted_at":"2024-12-20T08:07:11Z","abstract_excerpt":"Large Multimodal Models (LMMs) have demonstrated impressive performance across numerous academic benchmarks. However, fine-tuning still remains essential to achieve satisfactory performance on downstream tasks, while the task-specific tuning samples are usually not readily available or expensive and time-consuming to obtain. To address this, we propose an error-driven data-efficient tuning framework that aims to efficiently adapt generic LMMs to newly emerging tasks without requiring any task-specific training samples. In our approach, a generic LMM, acting as a student model, is first evaluat"},"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":"2412.15652","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-20T08:07:11Z","cross_cats_sorted":[],"title_canon_sha256":"d24ed4937fd6181c032e4533416ea484fd626b5bb366ef0242944749fd7622b6","abstract_canon_sha256":"a7535e467c13c96c45d12424c1543c54f41acd6fafc7a57fe924771d4093cb2e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:27.773185Z","signature_b64":"GDx0ILMKX9nYoZcoxfjY/hgUz8YIO9EwbvbQejFlHtFQOUaVxkJcBt2Vz2bNAGYTE8Q/8KdkreX1grrv52yfAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf7ed1037c1a3bb0b144788f438219855a9f537a52a3b1a6dc6857221941518a","last_reissued_at":"2026-07-05T09:52:27.772612Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:27.772612Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Error-driven Data-efficient Large Multimodal Model Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Barry Menglong Yao (UC Davis), Lifu Huang (UC Davis), Qifan Wang (Meta AI)","submitted_at":"2024-12-20T08:07:11Z","abstract_excerpt":"Large Multimodal Models (LMMs) have demonstrated impressive performance across numerous academic benchmarks. However, fine-tuning still remains essential to achieve satisfactory performance on downstream tasks, while the task-specific tuning samples are usually not readily available or expensive and time-consuming to obtain. To address this, we propose an error-driven data-efficient tuning framework that aims to efficiently adapt generic LMMs to newly emerging tasks without requiring any task-specific training samples. In our approach, a generic LMM, acting as a student model, is first evaluat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15652","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/2412.15652/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":"2412.15652","created_at":"2026-07-05T09:52:27.772697+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.15652v1","created_at":"2026-07-05T09:52:27.772697+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15652","created_at":"2026-07-05T09:52:27.772697+00:00"},{"alias_kind":"pith_short_12","alias_value":"X57NCA34DI53","created_at":"2026-07-05T09:52:27.772697+00:00"},{"alias_kind":"pith_short_16","alias_value":"X57NCA34DI53BMKE","created_at":"2026-07-05T09:52:27.772697+00:00"},{"alias_kind":"pith_short_8","alias_value":"X57NCA34","created_at":"2026-07-05T09:52:27.772697+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/X57NCA34DI53BMKEPCHUHAQZQV","json":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV.json","graph_json":"https://pith.science/api/pith-number/X57NCA34DI53BMKEPCHUHAQZQV/graph.json","events_json":"https://pith.science/api/pith-number/X57NCA34DI53BMKEPCHUHAQZQV/events.json","paper":"https://pith.science/paper/X57NCA34"},"agent_actions":{"view_html":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV","download_json":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV.json","view_paper":"https://pith.science/paper/X57NCA34","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.15652&json=true","fetch_graph":"https://pith.science/api/pith-number/X57NCA34DI53BMKEPCHUHAQZQV/graph.json","fetch_events":"https://pith.science/api/pith-number/X57NCA34DI53BMKEPCHUHAQZQV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV/action/storage_attestation","attest_author":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV/action/author_attestation","sign_citation":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV/action/citation_signature","submit_replication":"https://pith.science/pith/X57NCA34DI53BMKEPCHUHAQZQV/action/replication_record"}},"created_at":"2026-07-05T09:52:27.772697+00:00","updated_at":"2026-07-05T09:52:27.772697+00:00"}