{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4WDAUHA457DUW2WTSYB5FNEHEG","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":"5aaeead01686ab1b4c279cec546be484cd80a81db43f77b98d9bf6ed2bbda08f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-06T03:16:04Z","title_canon_sha256":"c554b01c096d1d3825197d7608432715ce7de78ed5eae3cdaa1cfa9a8615299a"},"schema_version":"1.0","source":{"id":"2602.07078","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.07078","created_at":"2026-06-23T02:13:20Z"},{"alias_kind":"arxiv_version","alias_value":"2602.07078v2","created_at":"2026-06-23T02:13:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.07078","created_at":"2026-06-23T02:13:20Z"},{"alias_kind":"pith_short_12","alias_value":"4WDAUHA457DU","created_at":"2026-06-23T02:13:20Z"},{"alias_kind":"pith_short_16","alias_value":"4WDAUHA457DUW2WT","created_at":"2026-06-23T02:13:20Z"},{"alias_kind":"pith_short_8","alias_value":"4WDAUHA4","created_at":"2026-06-23T02:13:20Z"}],"graph_snapshots":[{"event_id":"sha256:9331559155c4d990fa41d7cb70b4845b5f9a7b45f899bd28c5f84a930e21fd1d","target":"graph","created_at":"2026-06-23T02:13:20Z","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/2602.07078/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement Learning (RL) for Large Language Models (LLMs) often suffers from training collapse in long-horizon tasks due to exploding gradient variance. To mitigate this, a baseline is commonly introduced for advantage computation; however, traditional value models remain difficult to optimize, and standard group-based baselines overlook sequence heterogeneity. Although classic optimal baseline theory can achieve global variance reduction, it neglects token heterogeneity and requires prohibitive gradient-based computation. In this work, we derive the Optimal Token Baseline (OTB) from first ","authors_text":"Baoxiang Wang, Ge Zhang, Jiacai Liu, Jiawei Xu, Longtao Zheng, Qian Liu, Tianle Cai, Wei Liu, Yaxiang Zhang, Yingru Li, Yuxuan Tong, Zhenghai Xue, Ziniu Li","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-06T03:16:04Z","title":"The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.07078","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":"sha256:977f1c2707df547f9187e23514f1116a7f1ee1de78e342ba3adec50727ae67fe","target":"record","created_at":"2026-06-23T02:13:20Z","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":"5aaeead01686ab1b4c279cec546be484cd80a81db43f77b98d9bf6ed2bbda08f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-06T03:16:04Z","title_canon_sha256":"c554b01c096d1d3825197d7608432715ce7de78ed5eae3cdaa1cfa9a8615299a"},"schema_version":"1.0","source":{"id":"2602.07078","kind":"arxiv","version":2}},"canonical_sha256":"e5860a1c1cefc74b6ad39603d2b48721a989b33487196e8c602ffd747c0ba80b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e5860a1c1cefc74b6ad39603d2b48721a989b33487196e8c602ffd747c0ba80b","first_computed_at":"2026-06-23T02:13:20.781679Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T02:13:20.781679Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XagG2ZtjqCHj9K30nMzepMI2I8qjqDHG9gffiDeanGSZylXIVsP582QWFW9LXEds6fWwIdti6KPVUJSR1FepAA==","signature_status":"signed_v1","signed_at":"2026-06-23T02:13:20.782073Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.07078","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:977f1c2707df547f9187e23514f1116a7f1ee1de78e342ba3adec50727ae67fe","sha256:9331559155c4d990fa41d7cb70b4845b5f9a7b45f899bd28c5f84a930e21fd1d"],"state_sha256":"62f293d2fa1a73194830d7cd3e3d0ddc561cbd537ba840db6a2edc85141e8b93"}