{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OR3LRHENJJTB4ZBBTGPQM3P7SM","short_pith_number":"pith:OR3LRHEN","schema_version":"1.0","canonical_sha256":"7476b89c8d4a661e6421999f066dff93322cee3b65a15c96c43d398c31a599e3","source":{"kind":"arxiv","id":"2410.17961","version":2},"attestation_state":"computed","paper":{"title":"Closed-form merging of parameter-efficient modules for Federated Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jacopo Bonato, Luigi Sabetta, Matteo Mosconi, Pietro Buzzega, Riccardo Salami, Simone Calderara","submitted_at":"2024-10-23T15:30:13Z","abstract_excerpt":"Model merging has emerged as a crucial technique in Deep Learning, enabling the integration of multiple models into a unified system while preserving perfor-mance and scalability. In this respect, the compositional properties of low-rank adaptation techniques (e.g., LoRA) have proven beneficial, as simple averaging LoRA modules yields a single model that mostly integrates the capabilities of all individual modules. Building on LoRA, we take a step further by imposing that the merged model matches the responses of all learned modules. Solving this objective in closed form yields an indeterminat"},"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":"2410.17961","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-23T15:30:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e55d249761bce792c81f294472d7c12949a44f1275cf2d105fcdc336fe478f6a","abstract_canon_sha256":"06743c79c6f483fe2c1400c3438233823aa4f5b15aab9714926e191f9cdeba18"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:27:17.993864Z","signature_b64":"9Jf6c2JX20gNMsozk+hmHFlqHZuiUBhGtQB7q6jXTzErwpq/nbo3gCI412Djdk/gms1m4hmiEiW/Ns96cB5GCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7476b89c8d4a661e6421999f066dff93322cee3b65a15c96c43d398c31a599e3","last_reissued_at":"2026-07-05T10:27:17.993242Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:27:17.993242Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Closed-form merging of parameter-efficient modules for Federated Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jacopo Bonato, Luigi Sabetta, Matteo Mosconi, Pietro Buzzega, Riccardo Salami, Simone Calderara","submitted_at":"2024-10-23T15:30:13Z","abstract_excerpt":"Model merging has emerged as a crucial technique in Deep Learning, enabling the integration of multiple models into a unified system while preserving perfor-mance and scalability. In this respect, the compositional properties of low-rank adaptation techniques (e.g., LoRA) have proven beneficial, as simple averaging LoRA modules yields a single model that mostly integrates the capabilities of all individual modules. Building on LoRA, we take a step further by imposing that the merged model matches the responses of all learned modules. Solving this objective in closed form yields an indeterminat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.17961","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/2410.17961/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":"2410.17961","created_at":"2026-07-05T10:27:17.993310+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.17961v2","created_at":"2026-07-05T10:27:17.993310+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.17961","created_at":"2026-07-05T10:27:17.993310+00:00"},{"alias_kind":"pith_short_12","alias_value":"OR3LRHENJJTB","created_at":"2026-07-05T10:27:17.993310+00:00"},{"alias_kind":"pith_short_16","alias_value":"OR3LRHENJJTB4ZBB","created_at":"2026-07-05T10:27:17.993310+00:00"},{"alias_kind":"pith_short_8","alias_value":"OR3LRHEN","created_at":"2026-07-05T10:27:17.993310+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2505.12318","citing_title":"Task-agnostic Low-rank Residual Adaptation for Efficient Federated Continual Fine-Tuning","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2508.18187","citing_title":"BRAIN: Bias-Mitigation Continual Learning Approach to Vision-Brain Understanding","ref_index":100,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM","json":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM.json","graph_json":"https://pith.science/api/pith-number/OR3LRHENJJTB4ZBBTGPQM3P7SM/graph.json","events_json":"https://pith.science/api/pith-number/OR3LRHENJJTB4ZBBTGPQM3P7SM/events.json","paper":"https://pith.science/paper/OR3LRHEN"},"agent_actions":{"view_html":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM","download_json":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM.json","view_paper":"https://pith.science/paper/OR3LRHEN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.17961&json=true","fetch_graph":"https://pith.science/api/pith-number/OR3LRHENJJTB4ZBBTGPQM3P7SM/graph.json","fetch_events":"https://pith.science/api/pith-number/OR3LRHENJJTB4ZBBTGPQM3P7SM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM/action/storage_attestation","attest_author":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM/action/author_attestation","sign_citation":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM/action/citation_signature","submit_replication":"https://pith.science/pith/OR3LRHENJJTB4ZBBTGPQM3P7SM/action/replication_record"}},"created_at":"2026-07-05T10:27:17.993310+00:00","updated_at":"2026-07-05T10:27:17.993310+00:00"}