{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZBOGI4BIPLOZ7H55L6HBKXALZ2","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":"0e8614c9b359ffb2f7400e22e74b790b8876cbbfd3cacff03cbe942edc97a0a0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-06T15:48:26Z","title_canon_sha256":"4b0da46d97768c80333a8c11bada61d8fd84f8b0a7dd65091c69692a429c0e46"},"schema_version":"1.0","source":{"id":"2504.04524","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.04524","created_at":"2026-07-05T11:23:26Z"},{"alias_kind":"arxiv_version","alias_value":"2504.04524v2","created_at":"2026-07-05T11:23:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.04524","created_at":"2026-07-05T11:23:26Z"},{"alias_kind":"pith_short_12","alias_value":"ZBOGI4BIPLOZ","created_at":"2026-07-05T11:23:26Z"},{"alias_kind":"pith_short_16","alias_value":"ZBOGI4BIPLOZ7H55","created_at":"2026-07-05T11:23:26Z"},{"alias_kind":"pith_short_8","alias_value":"ZBOGI4BI","created_at":"2026-07-05T11:23:26Z"}],"graph_snapshots":[{"event_id":"sha256:271a376e789043654162e3c419336805815fb0225c42b993afd4664d163be09f","target":"graph","created_at":"2026-07-05T11:23:26Z","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/2504.04524/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recently, Large Language Models (LLMs) have rapidly evolved, approaching Artificial General Intelligence (AGI) while benefiting from large-scale reinforcement learning to enhance Human Alignment (HA) and Reasoning. Recent reward-based optimization algorithms, such as Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO) have achieved significant performance on reasoning tasks, whereas preference-based optimization algorithms such as Direct Preference Optimization (DPO) significantly improve the performance of LLMs on human alignment. However, despite the strong perfo","authors_text":"Guoqing Liu, Peiran Jin, Renqian Luo, Shufang Xie, Xuerui Su, Yingce Xia, Yue Wang, Yuting Liu, Zhiming Ma, Zun Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-06T15:48:26Z","title":"Trust Region Preference Approximation: A simple and stable reinforcement learning algorithm for LLM reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.04524","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:b152ce2ae017a0e19ab5b848bfe611b6e6011a49cdb4c081d80e7a5348cb3c74","target":"record","created_at":"2026-07-05T11:23:26Z","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":"0e8614c9b359ffb2f7400e22e74b790b8876cbbfd3cacff03cbe942edc97a0a0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-06T15:48:26Z","title_canon_sha256":"4b0da46d97768c80333a8c11bada61d8fd84f8b0a7dd65091c69692a429c0e46"},"schema_version":"1.0","source":{"id":"2504.04524","kind":"arxiv","version":2}},"canonical_sha256":"c85c6470287add9f9fbd5f8e155c0bcebdbcdc9c7a64e64789a13da9986e8a4c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c85c6470287add9f9fbd5f8e155c0bcebdbcdc9c7a64e64789a13da9986e8a4c","first_computed_at":"2026-07-05T11:23:26.707512Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:23:26.707512Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4kiQHxrRD8WqprjbLKOW1oVEXX7W7cShbPQsaG3qIDBNny9koTEJMzswW59BZUmWJv0KUVh7hXwg+iCq1C7ODw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:23:26.708009Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.04524","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b152ce2ae017a0e19ab5b848bfe611b6e6011a49cdb4c081d80e7a5348cb3c74","sha256:271a376e789043654162e3c419336805815fb0225c42b993afd4664d163be09f"],"state_sha256":"8d28bf1bb3e4c153dac938b4e6ec675f8966e317bbd6d27a856d53f857200762"}