{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XKX2DUEPUUWM7PHBZUMV4BH4FL","short_pith_number":"pith:XKX2DUEP","schema_version":"1.0","canonical_sha256":"baafa1d08fa52ccfbce1cd195e04fc2af1c78e084c09d0f007243ec24d2ef708","source":{"kind":"arxiv","id":"2409.13931","version":4},"attestation_state":"computed","paper":{"title":"On-Device Collaborative Language Modeling via a Mixture of Generalists and Specialists","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Bettina Messmer, Dongyang Fan, Martin Jaggi, Nikita Doikov","submitted_at":"2024-09-20T22:34:37Z","abstract_excerpt":"On-device LLMs have gained increasing attention for their ability to enhance privacy and provide a personalized user experience. To facilitate private learning with scarce data, Federated Learning has become a standard approach. However, it faces challenges such as computational resource heterogeneity and data heterogeneity among end users. We propose CoMiGS ($\\textbf{Co}$llaborative learning with a $\\textbf{Mi}$xture of $\\textbf{G}$eneralists and $\\textbf{S}$pecialists), the first approach to address both challenges. A key innovation of our method is the bi-level optimization formulation of t"},"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":"2409.13931","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-20T22:34:37Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"2f2c02e00aca7f0658b38a7c0e5d35f438ce9bb25b892e48250bb4525d9d693a","abstract_canon_sha256":"21deb8493b3f8262f4a927df21d4fe9e313aacf290c4cf5461fb0f3a7a1d7801"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:35.014930Z","signature_b64":"w1aByQKrdCNfJdUCq3U6moZYMrnal5+m12fVjoBRFppqt5DeE0JqOISQpPdmsllY9R9K9kgJdp66xepYaiuKBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"baafa1d08fa52ccfbce1cd195e04fc2af1c78e084c09d0f007243ec24d2ef708","last_reissued_at":"2026-07-05T11:11:35.014421Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:35.014421Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On-Device Collaborative Language Modeling via a Mixture of Generalists and Specialists","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Bettina Messmer, Dongyang Fan, Martin Jaggi, Nikita Doikov","submitted_at":"2024-09-20T22:34:37Z","abstract_excerpt":"On-device LLMs have gained increasing attention for their ability to enhance privacy and provide a personalized user experience. To facilitate private learning with scarce data, Federated Learning has become a standard approach. However, it faces challenges such as computational resource heterogeneity and data heterogeneity among end users. We propose CoMiGS ($\\textbf{Co}$llaborative learning with a $\\textbf{Mi}$xture of $\\textbf{G}$eneralists and $\\textbf{S}$pecialists), the first approach to address both challenges. A key innovation of our method is the bi-level optimization formulation of t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.13931","kind":"arxiv","version":4},"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/2409.13931/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":"2409.13931","created_at":"2026-07-05T11:11:35.014490+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.13931v4","created_at":"2026-07-05T11:11:35.014490+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.13931","created_at":"2026-07-05T11:11:35.014490+00:00"},{"alias_kind":"pith_short_12","alias_value":"XKX2DUEPUUWM","created_at":"2026-07-05T11:11:35.014490+00:00"},{"alias_kind":"pith_short_16","alias_value":"XKX2DUEPUUWM7PHB","created_at":"2026-07-05T11:11:35.014490+00:00"},{"alias_kind":"pith_short_8","alias_value":"XKX2DUEP","created_at":"2026-07-05T11:11:35.014490+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/XKX2DUEPUUWM7PHBZUMV4BH4FL","json":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL.json","graph_json":"https://pith.science/api/pith-number/XKX2DUEPUUWM7PHBZUMV4BH4FL/graph.json","events_json":"https://pith.science/api/pith-number/XKX2DUEPUUWM7PHBZUMV4BH4FL/events.json","paper":"https://pith.science/paper/XKX2DUEP"},"agent_actions":{"view_html":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL","download_json":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL.json","view_paper":"https://pith.science/paper/XKX2DUEP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.13931&json=true","fetch_graph":"https://pith.science/api/pith-number/XKX2DUEPUUWM7PHBZUMV4BH4FL/graph.json","fetch_events":"https://pith.science/api/pith-number/XKX2DUEPUUWM7PHBZUMV4BH4FL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL/action/storage_attestation","attest_author":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL/action/author_attestation","sign_citation":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL/action/citation_signature","submit_replication":"https://pith.science/pith/XKX2DUEPUUWM7PHBZUMV4BH4FL/action/replication_record"}},"created_at":"2026-07-05T11:11:35.014490+00:00","updated_at":"2026-07-05T11:11:35.014490+00:00"}