{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AGYIBW7WLYR2P7DQLQJ63ZNYMC","short_pith_number":"pith:AGYIBW7W","schema_version":"1.0","canonical_sha256":"01b080dbf65e23a7fc705c13ede5b860ba61e628919f284da432c834f48f4de0","source":{"kind":"arxiv","id":"2401.16265","version":1},"attestation_state":"computed","paper":{"title":"CO2: Efficient Distributed Training with Full Communication-Computation Overlap","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.CL","authors_text":"Dong Li, Shidi Li, Weigao Sun, Weixuan Sun, Xuyang Shen, Yiran Zhong, Yu Qiao, Zhen Qin","submitted_at":"2024-01-29T16:12:31Z","abstract_excerpt":"The fundamental success of large language models hinges upon the efficacious implementation of large-scale distributed training techniques. Nevertheless, building a vast, high-performance cluster featuring high-speed communication interconnectivity is prohibitively costly, and accessible only to prominent entities. In this work, we aim to lower this barrier and democratize large-scale training with limited bandwidth clusters. We propose a new approach called CO2 that introduces local-updating and asynchronous communication to the distributed data-parallel training, thereby facilitating the ful"},"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":"2401.16265","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-29T16:12:31Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"72e3f3995f54827e8bd69e95d8a3a23918580275433760af60e8d2ec61c9420e","abstract_canon_sha256":"efb77fef9790d218c092ec05a1000cc511a376be4a2a8b763f2db2f24447b28e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:38:46.761025Z","signature_b64":"tjDNJT284xv7ultZg+zi7yewuhM9Y/u2JRWXE/2vpR5Gtklr8D8YiUitnrz9Bd13xlO8tXj6nP3jEqHXBBSiDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"01b080dbf65e23a7fc705c13ede5b860ba61e628919f284da432c834f48f4de0","last_reissued_at":"2026-07-05T07:38:46.760601Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:38:46.760601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CO2: Efficient Distributed Training with Full Communication-Computation Overlap","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.CL","authors_text":"Dong Li, Shidi Li, Weigao Sun, Weixuan Sun, Xuyang Shen, Yiran Zhong, Yu Qiao, Zhen Qin","submitted_at":"2024-01-29T16:12:31Z","abstract_excerpt":"The fundamental success of large language models hinges upon the efficacious implementation of large-scale distributed training techniques. Nevertheless, building a vast, high-performance cluster featuring high-speed communication interconnectivity is prohibitively costly, and accessible only to prominent entities. In this work, we aim to lower this barrier and democratize large-scale training with limited bandwidth clusters. We propose a new approach called CO2 that introduces local-updating and asynchronous communication to the distributed data-parallel training, thereby facilitating the ful"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.16265","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/2401.16265/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":"2401.16265","created_at":"2026-07-05T07:38:46.760653+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.16265v1","created_at":"2026-07-05T07:38:46.760653+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.16265","created_at":"2026-07-05T07:38:46.760653+00:00"},{"alias_kind":"pith_short_12","alias_value":"AGYIBW7WLYR2","created_at":"2026-07-05T07:38:46.760653+00:00"},{"alias_kind":"pith_short_16","alias_value":"AGYIBW7WLYR2P7DQ","created_at":"2026-07-05T07:38:46.760653+00:00"},{"alias_kind":"pith_short_8","alias_value":"AGYIBW7W","created_at":"2026-07-05T07:38:46.760653+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.19169","citing_title":"Modeling the Impact of Fiber Latency on Compute-Communication Overlap in Geo-Distributed Multi-Datacenter AI Training","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09126","citing_title":"Cosine-Gated Adam-Decay: Drop-In Staleness-Aware Outer Optimization for Decoupled DiLoCo","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC","json":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC.json","graph_json":"https://pith.science/api/pith-number/AGYIBW7WLYR2P7DQLQJ63ZNYMC/graph.json","events_json":"https://pith.science/api/pith-number/AGYIBW7WLYR2P7DQLQJ63ZNYMC/events.json","paper":"https://pith.science/paper/AGYIBW7W"},"agent_actions":{"view_html":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC","download_json":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC.json","view_paper":"https://pith.science/paper/AGYIBW7W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.16265&json=true","fetch_graph":"https://pith.science/api/pith-number/AGYIBW7WLYR2P7DQLQJ63ZNYMC/graph.json","fetch_events":"https://pith.science/api/pith-number/AGYIBW7WLYR2P7DQLQJ63ZNYMC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC/action/storage_attestation","attest_author":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC/action/author_attestation","sign_citation":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC/action/citation_signature","submit_replication":"https://pith.science/pith/AGYIBW7WLYR2P7DQLQJ63ZNYMC/action/replication_record"}},"created_at":"2026-07-05T07:38:46.760653+00:00","updated_at":"2026-07-05T07:38:46.760653+00:00"}