{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:UMFUJZEM6KIZRLXWYJG6DJXCWB","short_pith_number":"pith:UMFUJZEM","canonical_record":{"source":{"id":"2410.23610","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-31T03:51:39Z","cross_cats_sorted":["cs.LG","math.ST","stat.TH"],"title_canon_sha256":"73d9d2663319dcf5f0bd6fa2e8fbbaac7ee2cb396e484dee23914da9af5eb213","abstract_canon_sha256":"fad3380e06c78a104ca5e83cc89edc8ba118c3ec6b094a7cf1bafbcd68b3b690"},"schema_version":"1.0"},"canonical_sha256":"a30b44e48cf29198aef6c24de1a6e2b0476ff96f688d08c6b48fb210e59c1d54","source":{"kind":"arxiv","id":"2410.23610","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.23610","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"arxiv_version","alias_value":"2410.23610v1","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.23610","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"pith_short_12","alias_value":"UMFUJZEM6KIZ","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"pith_short_16","alias_value":"UMFUJZEM6KIZRLXW","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"pith_short_8","alias_value":"UMFUJZEM","created_at":"2026-07-05T09:29:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:UMFUJZEM6KIZRLXWYJG6DJXCWB","target":"record","payload":{"canonical_record":{"source":{"id":"2410.23610","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-31T03:51:39Z","cross_cats_sorted":["cs.LG","math.ST","stat.TH"],"title_canon_sha256":"73d9d2663319dcf5f0bd6fa2e8fbbaac7ee2cb396e484dee23914da9af5eb213","abstract_canon_sha256":"fad3380e06c78a104ca5e83cc89edc8ba118c3ec6b094a7cf1bafbcd68b3b690"},"schema_version":"1.0"},"canonical_sha256":"a30b44e48cf29198aef6c24de1a6e2b0476ff96f688d08c6b48fb210e59c1d54","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:29:02.722777Z","signature_b64":"1KB3nR6Pz6Js0tTnmRzuzqQp25vOU37D0Im0JPgAHWTBik5RZytgLPjbJE9aY5WWleSQORITb612vNZf3AX1Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a30b44e48cf29198aef6c24de1a6e2b0476ff96f688d08c6b48fb210e59c1d54","last_reissued_at":"2026-07-05T09:29:02.722294Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:29:02.722294Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.23610","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:29:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2I3wmm1r2z203HdBbgDPdKM/ZsCkp1sqPR1mLNFEoOKTy/ztX9MileeM8BZfEhW3f8ioQ8/pxTSexs0AMlp7Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T03:09:13.412559Z"},"content_sha256":"23ecadf07d9ef668546661dc0ad67f27100939cb40d5163c524786b4a7c27f11","schema_version":"1.0","event_id":"sha256:23ecadf07d9ef668546661dc0ad67f27100939cb40d5163c524786b4a7c27f11"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:UMFUJZEM6KIZRLXWYJG6DJXCWB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Global Convergence in Training Large-Scale Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.ST","stat.TH"],"primary_cat":"stat.ML","authors_text":"Cheng Gao, Han Liu, Jason Matthew Klusowski, Jianqing Fan, Mengdi Wang, Yihan He, Yuan Cao, Zihao Li","submitted_at":"2024-10-31T03:51:39Z","abstract_excerpt":"Despite the widespread success of Transformers across various domains, their optimization guarantees in large-scale model settings are not well-understood. This paper rigorously analyzes the convergence properties of gradient flow in training Transformers with weight decay regularization. First, we construct the mean-field limit of large-scale Transformers, showing that as the model width and depth go to infinity, gradient flow converges to the Wasserstein gradient flow, which is represented by a partial differential equation. Then, we demonstrate that the gradient flow reaches a global minimu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.23610","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/2410.23610/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":"2026-07-05T09:29:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6l/oKlsd/H4CwkDWFcpLzDPoREaJW8Tfb3rqnopi6XznmwfRE/Svwh1wcjI0JkirMaAlYHGLHtr/KMM3Hm4FAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T03:09:13.413086Z"},"content_sha256":"5b8bd5c766c759d5b2b52b97bf827530b41ffcc4eab6788812ef0d4ac79b4c3d","schema_version":"1.0","event_id":"sha256:5b8bd5c766c759d5b2b52b97bf827530b41ffcc4eab6788812ef0d4ac79b4c3d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UMFUJZEM6KIZRLXWYJG6DJXCWB/bundle.json","state_url":"https://pith.science/pith/UMFUJZEM6KIZRLXWYJG6DJXCWB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UMFUJZEM6KIZRLXWYJG6DJXCWB/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":"2026-08-20T03:09:13Z","links":{"resolver":"https://pith.science/pith/UMFUJZEM6KIZRLXWYJG6DJXCWB","bundle":"https://pith.science/pith/UMFUJZEM6KIZRLXWYJG6DJXCWB/bundle.json","state":"https://pith.science/pith/UMFUJZEM6KIZRLXWYJG6DJXCWB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UMFUJZEM6KIZRLXWYJG6DJXCWB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UMFUJZEM6KIZRLXWYJG6DJXCWB","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":"fad3380e06c78a104ca5e83cc89edc8ba118c3ec6b094a7cf1bafbcd68b3b690","cross_cats_sorted":["cs.LG","math.ST","stat.TH"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-31T03:51:39Z","title_canon_sha256":"73d9d2663319dcf5f0bd6fa2e8fbbaac7ee2cb396e484dee23914da9af5eb213"},"schema_version":"1.0","source":{"id":"2410.23610","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.23610","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"arxiv_version","alias_value":"2410.23610v1","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.23610","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"pith_short_12","alias_value":"UMFUJZEM6KIZ","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"pith_short_16","alias_value":"UMFUJZEM6KIZRLXW","created_at":"2026-07-05T09:29:02Z"},{"alias_kind":"pith_short_8","alias_value":"UMFUJZEM","created_at":"2026-07-05T09:29:02Z"}],"graph_snapshots":[{"event_id":"sha256:5b8bd5c766c759d5b2b52b97bf827530b41ffcc4eab6788812ef0d4ac79b4c3d","target":"graph","created_at":"2026-07-05T09:29:02Z","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/2410.23610/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite the widespread success of Transformers across various domains, their optimization guarantees in large-scale model settings are not well-understood. This paper rigorously analyzes the convergence properties of gradient flow in training Transformers with weight decay regularization. First, we construct the mean-field limit of large-scale Transformers, showing that as the model width and depth go to infinity, gradient flow converges to the Wasserstein gradient flow, which is represented by a partial differential equation. Then, we demonstrate that the gradient flow reaches a global minimu","authors_text":"Cheng Gao, Han Liu, Jason Matthew Klusowski, Jianqing Fan, Mengdi Wang, Yihan He, Yuan Cao, Zihao Li","cross_cats":["cs.LG","math.ST","stat.TH"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-31T03:51:39Z","title":"Global Convergence in Training Large-Scale Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.23610","kind":"arxiv","version":1},"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":"sha256:23ecadf07d9ef668546661dc0ad67f27100939cb40d5163c524786b4a7c27f11","target":"record","created_at":"2026-07-05T09:29:02Z","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":"fad3380e06c78a104ca5e83cc89edc8ba118c3ec6b094a7cf1bafbcd68b3b690","cross_cats_sorted":["cs.LG","math.ST","stat.TH"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-31T03:51:39Z","title_canon_sha256":"73d9d2663319dcf5f0bd6fa2e8fbbaac7ee2cb396e484dee23914da9af5eb213"},"schema_version":"1.0","source":{"id":"2410.23610","kind":"arxiv","version":1}},"canonical_sha256":"a30b44e48cf29198aef6c24de1a6e2b0476ff96f688d08c6b48fb210e59c1d54","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a30b44e48cf29198aef6c24de1a6e2b0476ff96f688d08c6b48fb210e59c1d54","first_computed_at":"2026-07-05T09:29:02.722294Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:29:02.722294Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1KB3nR6Pz6Js0tTnmRzuzqQp25vOU37D0Im0JPgAHWTBik5RZytgLPjbJE9aY5WWleSQORITb612vNZf3AX1Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T09:29:02.722777Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.23610","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:23ecadf07d9ef668546661dc0ad67f27100939cb40d5163c524786b4a7c27f11","sha256:5b8bd5c766c759d5b2b52b97bf827530b41ffcc4eab6788812ef0d4ac79b4c3d"],"state_sha256":"f524f14316bb26e9c085780e8ad43a486c0fc1a9c5d51b71e6cfe1c6e675be44"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CHEvUWQQEvh9+Wjf2IzPK5cIKOLig/2qNSa2UpuTfPdzA8uAYwUUSpnqhj65DyfYlAObabY3XpMIiObLqMFJCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T03:09:13.419628Z","bundle_sha256":"1d639743bc053225efb5014bfdfcd0263bcbdc3aa9ed0fb3e6ba92d1db21ebd5"}}