{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZUUVVVXMOJIVWIZFEAXYLZR5US","short_pith_number":"pith:ZUUVVVXM","schema_version":"1.0","canonical_sha256":"cd295ad6ec72515b2325202f85e63da4895d246746f1cf145e975205002cb309","source":{"kind":"arxiv","id":"2401.06432","version":2},"attestation_state":"computed","paper":{"title":"Heterogeneous LoRA for Federated Fine-tuning of On-Device Foundation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Aldi Fahrezi, Gauri Joshi, Luyang Liu, Yae Jee Cho, Zheng Xu","submitted_at":"2024-01-12T07:52:07Z","abstract_excerpt":"Foundation models (FMs) adapt well to specific domains or tasks with fine-tuning, and federated learning (FL) enables the potential for privacy-preserving fine-tuning of the FMs with on-device local data. For federated fine-tuning of FMs, we consider the FMs with small to medium parameter sizes of single digit billion at maximum, referred to as on-device FMs (ODFMs) that can be deployed on devices for inference but can only be fine-tuned with parameter efficient methods. In our work, we tackle the data and system heterogeneity problem of federated fine-tuning of ODFMs by proposing a novel meth"},"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":"2401.06432","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-12T07:52:07Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"bde51ff40c14c5b72cf84f3378943a924fbdd22b674a2cb4e2d27d87862b3a90","abstract_canon_sha256":"223fc60c9f846f69a7d7b35eb7b4fd90a43105c055b893d0f06b77c528cab5ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:47:27.990123Z","signature_b64":"7CgXwy0XzaWdDqGUpWmN0VP1MCLQw46UK4EnBG3gx5fyMC6FtP2OrToaDk7QZPjGj5bIJNKuj2HNbnBENa9QDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd295ad6ec72515b2325202f85e63da4895d246746f1cf145e975205002cb309","last_reissued_at":"2026-07-05T07:47:27.989607Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:47:27.989607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Heterogeneous LoRA for Federated Fine-tuning of On-Device Foundation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Aldi Fahrezi, Gauri Joshi, Luyang Liu, Yae Jee Cho, Zheng Xu","submitted_at":"2024-01-12T07:52:07Z","abstract_excerpt":"Foundation models (FMs) adapt well to specific domains or tasks with fine-tuning, and federated learning (FL) enables the potential for privacy-preserving fine-tuning of the FMs with on-device local data. For federated fine-tuning of FMs, we consider the FMs with small to medium parameter sizes of single digit billion at maximum, referred to as on-device FMs (ODFMs) that can be deployed on devices for inference but can only be fine-tuned with parameter efficient methods. In our work, we tackle the data and system heterogeneity problem of federated fine-tuning of ODFMs by proposing a novel meth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.06432","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/2401.06432/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":"2401.06432","created_at":"2026-07-05T07:47:27.989670+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.06432v2","created_at":"2026-07-05T07:47:27.989670+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.06432","created_at":"2026-07-05T07:47:27.989670+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZUUVVVXMOJIV","created_at":"2026-07-05T07:47:27.989670+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZUUVVVXMOJIVWIZF","created_at":"2026-07-05T07:47:27.989670+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZUUVVVXM","created_at":"2026-07-05T07:47:27.989670+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07529","citing_title":"FedMark-FM: Auditable, Risk-Adjusted Data Markets for Federated Foundation-Model Adaptation","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11857","citing_title":"Beyond Parameter Aggregation: Semantic Consensus for Federated Fine-Tuning of LLMs","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10272","citing_title":"DP-LAC: Lightweight Adaptive Clipping for Differentially Private Federated Fine-tuning of Language Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05769","citing_title":"Adaptive Selection of LoRA Components in Privacy-Preserving Federated Learning","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06297","citing_title":"FedSpy-LLM: Towards Scalable and Generalizable Data Reconstruction Attacks from Gradients on LLMs","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US","json":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US.json","graph_json":"https://pith.science/api/pith-number/ZUUVVVXMOJIVWIZFEAXYLZR5US/graph.json","events_json":"https://pith.science/api/pith-number/ZUUVVVXMOJIVWIZFEAXYLZR5US/events.json","paper":"https://pith.science/paper/ZUUVVVXM"},"agent_actions":{"view_html":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US","download_json":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US.json","view_paper":"https://pith.science/paper/ZUUVVVXM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.06432&json=true","fetch_graph":"https://pith.science/api/pith-number/ZUUVVVXMOJIVWIZFEAXYLZR5US/graph.json","fetch_events":"https://pith.science/api/pith-number/ZUUVVVXMOJIVWIZFEAXYLZR5US/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US/action/storage_attestation","attest_author":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US/action/author_attestation","sign_citation":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US/action/citation_signature","submit_replication":"https://pith.science/pith/ZUUVVVXMOJIVWIZFEAXYLZR5US/action/replication_record"}},"created_at":"2026-07-05T07:47:27.989670+00:00","updated_at":"2026-07-05T07:47:27.989670+00:00"}