{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:FW6K2W2XWJUVXIE7X3FNH5TNZ5","short_pith_number":"pith:FW6K2W2X","canonical_record":{"source":{"id":"2411.17866","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T20:31:11Z","cross_cats_sorted":[],"title_canon_sha256":"65204abf8f27e5e8bf228e2f65eb6180986e153eb29678972f4ec2f2b3725766","abstract_canon_sha256":"0891f838c80b548ffc50cac6b8b450a943197a90b2efefa7c5c7bb5f22e72e61"},"schema_version":"1.0"},"canonical_sha256":"2dbcad5b57b2695ba09fbecad3f66dcf477412d527aff732d12cd14bdf5c3700","source":{"kind":"arxiv","id":"2411.17866","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.17866","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"arxiv_version","alias_value":"2411.17866v2","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17866","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"pith_short_12","alias_value":"FW6K2W2XWJUV","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"pith_short_16","alias_value":"FW6K2W2XWJUVXIE7","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"pith_short_8","alias_value":"FW6K2W2X","created_at":"2026-07-05T10:26:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:FW6K2W2XWJUVXIE7X3FNH5TNZ5","target":"record","payload":{"canonical_record":{"source":{"id":"2411.17866","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T20:31:11Z","cross_cats_sorted":[],"title_canon_sha256":"65204abf8f27e5e8bf228e2f65eb6180986e153eb29678972f4ec2f2b3725766","abstract_canon_sha256":"0891f838c80b548ffc50cac6b8b450a943197a90b2efefa7c5c7bb5f22e72e61"},"schema_version":"1.0"},"canonical_sha256":"2dbcad5b57b2695ba09fbecad3f66dcf477412d527aff732d12cd14bdf5c3700","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:26:41.738028Z","signature_b64":"3nPjFXrsXcJ3Z9V4ur0h5fzBlPX934DnH4nmc3x30Lck9SFuCxIKWORqfl+BTc+8bl3+lXrnUqTuZHcEZZghBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2dbcad5b57b2695ba09fbecad3f66dcf477412d527aff732d12cd14bdf5c3700","last_reissued_at":"2026-07-05T10:26:41.737409Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:26:41.737409Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.17866","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:26:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4K8w6DzWYojpn6twEyaRIAeBmeNV9Kbk8u82oA7J5p6MEKHiIf0wrLaXJUnK4Fw5CN4Cd8NhMpMNcPA2wuJ0DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T21:18:27.151605Z"},"content_sha256":"cd02bc489cf184e6745272965c439b36e60da6cf5598edd4eefc8ab2e299ddd1","schema_version":"1.0","event_id":"sha256:cd02bc489cf184e6745272965c439b36e60da6cf5598edd4eefc8ab2e299ddd1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:FW6K2W2XWJUVXIE7X3FNH5TNZ5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Distributed Sign Momentum with Local Steps for Training Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"An Xu, Cong Xie, Ding Zhou, Shuhua Yu, Soummya Kar, Xin Liu, Zhi Zhang","submitted_at":"2024-11-26T20:31:11Z","abstract_excerpt":"Pre-training Transformer models is resource-intensive, and recent studies have shown that sign momentum is an efficient technique for training large-scale deep learning models, particularly Transformers. However, its application in distributed training remains underexplored. This paper investigates a novel communication-efficient distributed sign momentum method with multiple local steps, to cope with the scenarios where communicating at every step is prohibitive. Our proposed method allows for a broad class of base optimizers for local steps, and uses sign momentum in the global step, where m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17866","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/2411.17866/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:26:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GX6RRkviMXI1roAo7d7YD9CDFe51PF/gRlgTL1ZVp+ky5q2z7iCduN0H87HJ9rY2RdkXxbd3pKAbOIBogZ6ZDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T21:18:27.152095Z"},"content_sha256":"fc83b0eb1f20c2be9bf28a923146b2c8c28366b69c3d50314260be83d7a4b25c","schema_version":"1.0","event_id":"sha256:fc83b0eb1f20c2be9bf28a923146b2c8c28366b69c3d50314260be83d7a4b25c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FW6K2W2XWJUVXIE7X3FNH5TNZ5/bundle.json","state_url":"https://pith.science/pith/FW6K2W2XWJUVXIE7X3FNH5TNZ5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FW6K2W2XWJUVXIE7X3FNH5TNZ5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-16T21:18:27Z","links":{"resolver":"https://pith.science/pith/FW6K2W2XWJUVXIE7X3FNH5TNZ5","bundle":"https://pith.science/pith/FW6K2W2XWJUVXIE7X3FNH5TNZ5/bundle.json","state":"https://pith.science/pith/FW6K2W2XWJUVXIE7X3FNH5TNZ5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FW6K2W2XWJUVXIE7X3FNH5TNZ5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:FW6K2W2XWJUVXIE7X3FNH5TNZ5","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"0891f838c80b548ffc50cac6b8b450a943197a90b2efefa7c5c7bb5f22e72e61","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T20:31:11Z","title_canon_sha256":"65204abf8f27e5e8bf228e2f65eb6180986e153eb29678972f4ec2f2b3725766"},"schema_version":"1.0","source":{"id":"2411.17866","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.17866","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"arxiv_version","alias_value":"2411.17866v2","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17866","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"pith_short_12","alias_value":"FW6K2W2XWJUV","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"pith_short_16","alias_value":"FW6K2W2XWJUVXIE7","created_at":"2026-07-05T10:26:41Z"},{"alias_kind":"pith_short_8","alias_value":"FW6K2W2X","created_at":"2026-07-05T10:26:41Z"}],"graph_snapshots":[{"event_id":"sha256:fc83b0eb1f20c2be9bf28a923146b2c8c28366b69c3d50314260be83d7a4b25c","target":"graph","created_at":"2026-07-05T10:26:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2411.17866/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pre-training Transformer models is resource-intensive, and recent studies have shown that sign momentum is an efficient technique for training large-scale deep learning models, particularly Transformers. However, its application in distributed training remains underexplored. This paper investigates a novel communication-efficient distributed sign momentum method with multiple local steps, to cope with the scenarios where communicating at every step is prohibitive. Our proposed method allows for a broad class of base optimizers for local steps, and uses sign momentum in the global step, where m","authors_text":"An Xu, Cong Xie, Ding Zhou, Shuhua Yu, Soummya Kar, Xin Liu, Zhi Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T20:31:11Z","title":"Distributed Sign Momentum with Local Steps for Training Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17866","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:cd02bc489cf184e6745272965c439b36e60da6cf5598edd4eefc8ab2e299ddd1","target":"record","created_at":"2026-07-05T10:26:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"0891f838c80b548ffc50cac6b8b450a943197a90b2efefa7c5c7bb5f22e72e61","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-26T20:31:11Z","title_canon_sha256":"65204abf8f27e5e8bf228e2f65eb6180986e153eb29678972f4ec2f2b3725766"},"schema_version":"1.0","source":{"id":"2411.17866","kind":"arxiv","version":2}},"canonical_sha256":"2dbcad5b57b2695ba09fbecad3f66dcf477412d527aff732d12cd14bdf5c3700","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2dbcad5b57b2695ba09fbecad3f66dcf477412d527aff732d12cd14bdf5c3700","first_computed_at":"2026-07-05T10:26:41.737409Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:26:41.737409Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3nPjFXrsXcJ3Z9V4ur0h5fzBlPX934DnH4nmc3x30Lck9SFuCxIKWORqfl+BTc+8bl3+lXrnUqTuZHcEZZghBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:26:41.738028Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.17866","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cd02bc489cf184e6745272965c439b36e60da6cf5598edd4eefc8ab2e299ddd1","sha256:fc83b0eb1f20c2be9bf28a923146b2c8c28366b69c3d50314260be83d7a4b25c"],"state_sha256":"4be27f573b71b073c4bb1a648bbcb8bd3f206563c8d20df0a953f48ff99d0c76"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9kagQ3taznZ3IyQhLtGutDXDb6FO+wE7fVbHL373G2EiiqTsvBOdqZSCrr3utUTCmaXSuU8XsnZeEBT1lc9FAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T21:18:27.155244Z","bundle_sha256":"2e99722704a70d537a97b9cd37c8c70b3ae20b8f093b6d017c69c1f9b022d160"}}