{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UDPV3RHUGBQVHLKZITY4TOCAJ6","short_pith_number":"pith:UDPV3RHU","schema_version":"1.0","canonical_sha256":"a0df5dc4f4306153ad5944f1c9b8404f9ba298edc88f140d92c4add88cbe11b7","source":{"kind":"arxiv","id":"2501.13985","version":1},"attestation_state":"computed","paper":{"title":"Pilot: Building the Federated Multimodal Instruction Tuning Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Baochen Xiong, Changsheng Xu, Xiaoshan Yang, Yaguang Song, Yaowei Wang","submitted_at":"2025-01-23T07:49:24Z","abstract_excerpt":"In this paper, we explore a novel federated multimodal instruction tuning task(FedMIT), which is significant for collaboratively fine-tuning MLLMs on different types of multimodal instruction data on distributed devices. To solve the new task, we propose a federated multimodal instruction tuning framework(Pilot). Our framework integrates two stages of \"adapter on adapter\" into the connector of the vision encoder and the LLM. In stage 1, we extract task-specific features and client-specific features from visual information. In stage 2, we build the cross-task Mixture-of-Adapters(CT-MoA) module "},"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":"2501.13985","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-23T07:49:24Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"dd3acacc2bc1ba496a7f48b37e0f3b56314599ee50d3c9d13265d969dbc03a33","abstract_canon_sha256":"4fce2e5175349baf1b9cc10eea55af7426f18f355f81211893b5799720aa6af6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:43.467077Z","signature_b64":"WrSumabBYJZvnxLV7Tm59QBkRJA3/ybrDwm+gpMWpUQX4BE61IgfalGCoDB3ZogPHum2TjaJZH5uqvpMw8VeDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a0df5dc4f4306153ad5944f1c9b8404f9ba298edc88f140d92c4add88cbe11b7","last_reissued_at":"2026-07-05T10:04:43.466676Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:43.466676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pilot: Building the Federated Multimodal Instruction Tuning Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Baochen Xiong, Changsheng Xu, Xiaoshan Yang, Yaguang Song, Yaowei Wang","submitted_at":"2025-01-23T07:49:24Z","abstract_excerpt":"In this paper, we explore a novel federated multimodal instruction tuning task(FedMIT), which is significant for collaboratively fine-tuning MLLMs on different types of multimodal instruction data on distributed devices. To solve the new task, we propose a federated multimodal instruction tuning framework(Pilot). Our framework integrates two stages of \"adapter on adapter\" into the connector of the vision encoder and the LLM. In stage 1, we extract task-specific features and client-specific features from visual information. In stage 2, we build the cross-task Mixture-of-Adapters(CT-MoA) module "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.13985","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/2501.13985/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":"2501.13985","created_at":"2026-07-05T10:04:43.466736+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.13985v1","created_at":"2026-07-05T10:04:43.466736+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.13985","created_at":"2026-07-05T10:04:43.466736+00:00"},{"alias_kind":"pith_short_12","alias_value":"UDPV3RHUGBQV","created_at":"2026-07-05T10:04:43.466736+00:00"},{"alias_kind":"pith_short_16","alias_value":"UDPV3RHUGBQVHLKZ","created_at":"2026-07-05T10:04:43.466736+00:00"},{"alias_kind":"pith_short_8","alias_value":"UDPV3RHU","created_at":"2026-07-05T10:04:43.466736+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16690","citing_title":"UB-SMoE: Universally Balanced Sparse Mixture-of-Experts for Resource-adaptive Federated Fine-tuning of Foundation Models","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6","json":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6.json","graph_json":"https://pith.science/api/pith-number/UDPV3RHUGBQVHLKZITY4TOCAJ6/graph.json","events_json":"https://pith.science/api/pith-number/UDPV3RHUGBQVHLKZITY4TOCAJ6/events.json","paper":"https://pith.science/paper/UDPV3RHU"},"agent_actions":{"view_html":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6","download_json":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6.json","view_paper":"https://pith.science/paper/UDPV3RHU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.13985&json=true","fetch_graph":"https://pith.science/api/pith-number/UDPV3RHUGBQVHLKZITY4TOCAJ6/graph.json","fetch_events":"https://pith.science/api/pith-number/UDPV3RHUGBQVHLKZITY4TOCAJ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6/action/storage_attestation","attest_author":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6/action/author_attestation","sign_citation":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6/action/citation_signature","submit_replication":"https://pith.science/pith/UDPV3RHUGBQVHLKZITY4TOCAJ6/action/replication_record"}},"created_at":"2026-07-05T10:04:43.466736+00:00","updated_at":"2026-07-05T10:04:43.466736+00:00"}