{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2VIZR77Y7OL4I324ZXCUGLPSR4","short_pith_number":"pith:2VIZR77Y","schema_version":"1.0","canonical_sha256":"d55198fff8fb97c46f5ccdc5432df28f2cb2954200700e3dc23210efe575f5f2","source":{"kind":"arxiv","id":"2608.07974","version":1},"attestation_state":"computed","paper":{"title":"ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Chao Huang, Ming Tang, Wentao Dai, Xuanran Li, Yuxiang Zhang","submitted_at":"2026-08-08T07:19:21Z","abstract_excerpt":"Large language model (LLM) fine-tuning at the edge adapts the model to scenario-specific data while preserving privacy. Although existing studies proposed pipeline parallelism to address the limited memory and computing resources of edge devices, they commonly rely on backpropagation (BP) training, which has a fundamental limitation of update locking and could experience severe throughput and memory bottlenecks. In this work, we propose a BP-free algorithm, called ZeroLock, that decouples the model updates into independent chunk updates by local objective construction. It breaks the update loc"},"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":"2608.07974","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-08-08T07:19:21Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"a34f12e25ebe2ef8fa0ffab8d6c5dec52d5a05a755585d72fa1d20333d34a14a","abstract_canon_sha256":"9a3f87aba1795b5ec160a139b28e709b5a588555988a8390aa55dab1b0878e2a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T01:21:16.974263Z","signature_b64":"fMidqj9hDzAr9Drego0/13gFC0etL+/D/huIRnwiUhIBTXfUUNYkYwxtxshoJHfHkrb5LFBT8zHzKy6u8nNnDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d55198fff8fb97c46f5ccdc5432df28f2cb2954200700e3dc23210efe575f5f2","last_reissued_at":"2026-08-11T01:21:16.971602Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T01:21:16.971602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Chao Huang, Ming Tang, Wentao Dai, Xuanran Li, Yuxiang Zhang","submitted_at":"2026-08-08T07:19:21Z","abstract_excerpt":"Large language model (LLM) fine-tuning at the edge adapts the model to scenario-specific data while preserving privacy. Although existing studies proposed pipeline parallelism to address the limited memory and computing resources of edge devices, they commonly rely on backpropagation (BP) training, which has a fundamental limitation of update locking and could experience severe throughput and memory bottlenecks. In this work, we propose a BP-free algorithm, called ZeroLock, that decouples the model updates into independent chunk updates by local objective construction. It breaks the update loc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07974","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/2608.07974/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":"2608.07974","created_at":"2026-08-11T01:21:16.972385+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.07974v1","created_at":"2026-08-11T01:21:16.972385+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07974","created_at":"2026-08-11T01:21:16.972385+00:00"},{"alias_kind":"pith_short_12","alias_value":"2VIZR77Y7OL4","created_at":"2026-08-11T01:21:16.972385+00:00"},{"alias_kind":"pith_short_16","alias_value":"2VIZR77Y7OL4I324","created_at":"2026-08-11T01:21:16.972385+00:00"},{"alias_kind":"pith_short_8","alias_value":"2VIZR77Y","created_at":"2026-08-11T01:21:16.972385+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/2VIZR77Y7OL4I324ZXCUGLPSR4","json":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4.json","graph_json":"https://pith.science/api/pith-number/2VIZR77Y7OL4I324ZXCUGLPSR4/graph.json","events_json":"https://pith.science/api/pith-number/2VIZR77Y7OL4I324ZXCUGLPSR4/events.json","paper":"https://pith.science/paper/2VIZR77Y"},"agent_actions":{"view_html":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4","download_json":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4.json","view_paper":"https://pith.science/paper/2VIZR77Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.07974&json=true","fetch_graph":"https://pith.science/api/pith-number/2VIZR77Y7OL4I324ZXCUGLPSR4/graph.json","fetch_events":"https://pith.science/api/pith-number/2VIZR77Y7OL4I324ZXCUGLPSR4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4/action/storage_attestation","attest_author":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4/action/author_attestation","sign_citation":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4/action/citation_signature","submit_replication":"https://pith.science/pith/2VIZR77Y7OL4I324ZXCUGLPSR4/action/replication_record"}},"created_at":"2026-08-11T01:21:16.972385+00:00","updated_at":"2026-08-11T01:21:16.972385+00:00"}