{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FUOBIBQTY3LQWF664ALUXCQMZZ","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":"f38ccb565f8589f7b01bd058cb1f495187ce6b7f82bc5e86749732cc303600bb","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-26T10:06:37Z","title_canon_sha256":"f52c51631746ebef814a84733e543ce62fb04f61c99b7810860afd17f5f8c0b1"},"schema_version":"1.0","source":{"id":"2502.19002","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.19002","created_at":"2026-07-05T11:20:51Z"},{"alias_kind":"arxiv_version","alias_value":"2502.19002v2","created_at":"2026-07-05T11:20:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.19002","created_at":"2026-07-05T11:20:51Z"},{"alias_kind":"pith_short_12","alias_value":"FUOBIBQTY3LQ","created_at":"2026-07-05T11:20:51Z"},{"alias_kind":"pith_short_16","alias_value":"FUOBIBQTY3LQWF66","created_at":"2026-07-05T11:20:51Z"},{"alias_kind":"pith_short_8","alias_value":"FUOBIBQT","created_at":"2026-07-05T11:20:51Z"}],"graph_snapshots":[{"event_id":"sha256:341ba507460fe24172989852711a9847c1df33d27ddddacae5f80659036d56e0","target":"graph","created_at":"2026-07-05T11:20:51Z","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/2502.19002/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformers consist of diverse building blocks, such as embedding layers, normalization layers, self-attention mechanisms, and point-wise feedforward networks. Thus, understanding the differences and interactions among these blocks is important. In this paper, we uncover a clear Sharpness Disparity across these blocks, which emerges early in training and intriguingly persists throughout the training process. Motivated by this finding, we propose Blockwise Learning Rate (LR), a strategy that tailors the LR to each block's sharpness, accelerating large language model (LLM) pre-training. By inte","authors_text":"Jinbo Wang, Junchi Yan, Lei Wu, Mingze Wang, Weinan E, Zhanpeng Zhou","cross_cats":["cs.AI","math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-26T10:06:37Z","title":"The Sharpness Disparity Principle in Transformers for Accelerating Language Model Pre-Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.19002","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":"sha256:0cfb81467d8e58b5f53535d3d51cf20f6c25d920061e480e364ff725475611dd","target":"record","created_at":"2026-07-05T11:20:51Z","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":"f38ccb565f8589f7b01bd058cb1f495187ce6b7f82bc5e86749732cc303600bb","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-26T10:06:37Z","title_canon_sha256":"f52c51631746ebef814a84733e543ce62fb04f61c99b7810860afd17f5f8c0b1"},"schema_version":"1.0","source":{"id":"2502.19002","kind":"arxiv","version":2}},"canonical_sha256":"2d1c140613c6d70b17dee0174b8a0cce785c95d679b6c6cfd2b3629e9f821f24","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2d1c140613c6d70b17dee0174b8a0cce785c95d679b6c6cfd2b3629e9f821f24","first_computed_at":"2026-07-05T11:20:51.213430Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:20:51.213430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zNXUdlDNsSYDRY/IAZDrRT7W9IioCUS47HynpdkGvYNCeM2k9dTgJa7lvu1HJ8kzfQQzv1uzfZVT6oc1QhJqBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:20:51.213893Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.19002","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0cfb81467d8e58b5f53535d3d51cf20f6c25d920061e480e364ff725475611dd","sha256:341ba507460fe24172989852711a9847c1df33d27ddddacae5f80659036d56e0"],"state_sha256":"68ac2dcfdc9c5619c28b8cc8af995dcb0211803cecc2e43b5f820da2c79bbc88"}