{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:3JKBJ6XXKLPWAYFEI2OINYRHTN","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":"a2ec1e5d386e9cf493d836e8580907fce3e18c9884a5234937815862bc6da1bc","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T06:49:57Z","title_canon_sha256":"16f073554ab6719c55fdcc2ee950d35597d354033ff95305a24b9ccc10506856"},"schema_version":"1.0","source":{"id":"2505.24275","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24275","created_at":"2026-05-21T01:04:14Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24275v3","created_at":"2026-05-21T01:04:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24275","created_at":"2026-05-21T01:04:14Z"},{"alias_kind":"pith_short_12","alias_value":"3JKBJ6XXKLPW","created_at":"2026-05-21T01:04:14Z"},{"alias_kind":"pith_short_16","alias_value":"3JKBJ6XXKLPWAYFE","created_at":"2026-05-21T01:04:14Z"},{"alias_kind":"pith_short_8","alias_value":"3JKBJ6XX","created_at":"2026-05-21T01:04:14Z"}],"graph_snapshots":[{"event_id":"sha256:8a0e42e9d4edfd916230fbdeded5879562c782f04c8f83b0955ee735a14240dc","target":"graph","created_at":"2026-05-21T01:04:14Z","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/2505.24275/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose GradPower, a lightweight gradient-transformation technique for accelerating language model pre-training. Given a gradient vector $g=(g_i)_i$, GradPower first applies the elementwise sign-power transformation: $\\varphi_p(g)=({\\rm sign}(g_i)|g_i|^p)_{i}$ for a fixed $p>0$, and then feeds the transformed gradient into a base optimizer. Notably, GradPower requires only a single-line code change and no modifications to the base optimizer's internal logic, including the hyperparameters. When applied to Adam (termed AdamPower), GradPower consistently achieves lower terminal loss across div","authors_text":"Jiaqi Zhang, Jinbo Wang, Lei Wu, Mingze Wang, Peng Pei, Weinan E, Wei Wang, Xunliang Cai","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T06:49:57Z","title":"GradPower: Powering Gradients for Faster Language Model Pre-Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24275","kind":"arxiv","version":3},"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":"sha256:369937242a1fc3fdbf3e42dd8ad3b72877c8aa6f06dcf8ec36900cee4397b1bc","target":"record","created_at":"2026-05-21T01:04:14Z","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":"a2ec1e5d386e9cf493d836e8580907fce3e18c9884a5234937815862bc6da1bc","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T06:49:57Z","title_canon_sha256":"16f073554ab6719c55fdcc2ee950d35597d354033ff95305a24b9ccc10506856"},"schema_version":"1.0","source":{"id":"2505.24275","kind":"arxiv","version":3}},"canonical_sha256":"da5414faf752df6060a4469c86e2279b48773390bd662160c64ad6e6c848e052","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da5414faf752df6060a4469c86e2279b48773390bd662160c64ad6e6c848e052","first_computed_at":"2026-05-21T01:04:14.251596Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:04:14.251596Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OeGbVwBQ2exhfRzzvFxnapp2YWZxs/MzI42EPVHk604uQCkHXVtl4Y0QSzN22zTB4KjImuPbXWP9ZxYMnxFpDw==","signature_status":"signed_v1","signed_at":"2026-05-21T01:04:14.252386Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.24275","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:369937242a1fc3fdbf3e42dd8ad3b72877c8aa6f06dcf8ec36900cee4397b1bc","sha256:8a0e42e9d4edfd916230fbdeded5879562c782f04c8f83b0955ee735a14240dc"],"state_sha256":"a5f3fab0691e0d086e630326ab8ef4c4d8ed213ef75ac45d546c75568c4542ac"}