{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZOJ2MBIOGIJJFKPZEUK2IIEDNI","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":"f25ac7aeb028edabea2d6f920174b306f5a269f3f94bfa24395a6cf4efcad47a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T18:59:26Z","title_canon_sha256":"333922b3079274129d078e62fc5af49b3b4af0a578e9b1f299391b2d88ef52ee"},"schema_version":"1.0","source":{"id":"2605.20402","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.20402","created_at":"2026-05-21T01:04:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.20402v1","created_at":"2026-05-21T01:04:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20402","created_at":"2026-05-21T01:04:36Z"},{"alias_kind":"pith_short_12","alias_value":"ZOJ2MBIOGIJJ","created_at":"2026-05-21T01:04:36Z"},{"alias_kind":"pith_short_16","alias_value":"ZOJ2MBIOGIJJFKPZ","created_at":"2026-05-21T01:04:36Z"},{"alias_kind":"pith_short_8","alias_value":"ZOJ2MBIO","created_at":"2026-05-21T01:04:36Z"}],"graph_snapshots":[{"event_id":"sha256:d70df14d00b9420ef733e1076fbe4b129f773aa061fa2b7b5ad9ac8c8e76f060","target":"graph","created_at":"2026-05-21T01:04:36Z","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/2605.20402/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"MXFP4 arithmetic can dramatically accelerate reinforcement learning (RL) post-training of large language models (LLMs), yet the quantization error introduces severe accuracy degradation. Existing work treats the quantization error as a monolithic noise term, missing the distinct mechanisms upon interpreting how quantization error damages training. We prove an exact three-way decomposition of quantization error and show how each component dominates a distinct RL training pathway. Our theoretical and empirical analysis decomposes the MXFP4 quantization error into three additive components: \"scal","authors_text":"Shiliang Wu, Xiaocan Li, Zheng Shen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T18:59:26Z","title":"Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20402","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":"sha256:2b8bb78982b19eb2f5684117b4449e6e4014f0d8c7a66ec18244da7752588f32","target":"record","created_at":"2026-05-21T01:04:36Z","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":"f25ac7aeb028edabea2d6f920174b306f5a269f3f94bfa24395a6cf4efcad47a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T18:59:26Z","title_canon_sha256":"333922b3079274129d078e62fc5af49b3b4af0a578e9b1f299391b2d88ef52ee"},"schema_version":"1.0","source":{"id":"2605.20402","kind":"arxiv","version":1}},"canonical_sha256":"cb93a6050e321292a9f92515a420836a1bdc38e79540e0f69978e8d2e4aaf322","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb93a6050e321292a9f92515a420836a1bdc38e79540e0f69978e8d2e4aaf322","first_computed_at":"2026-05-21T01:04:36.635515Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:04:36.635515Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sZ22Z9nZ5ruBi0yyg6kQkbr3Fc1W0oaU0a3UALcZGJCofRtK4OodkT2+xZhWbKQOZgro1Z+DuSeI96krU47lAQ==","signature_status":"signed_v1","signed_at":"2026-05-21T01:04:36.635947Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.20402","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2b8bb78982b19eb2f5684117b4449e6e4014f0d8c7a66ec18244da7752588f32","sha256:d70df14d00b9420ef733e1076fbe4b129f773aa061fa2b7b5ad9ac8c8e76f060"],"state_sha256":"cd6f539677a249465c8ce263829b6176a6d985f83a85b2d62e27c19fe90e77f5"}