{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:L5HEIOCSO3XSXF4FM4TBFXCEZN","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":"0814c105be5c38a9e76396cc46d681d73185561b2524c122fa2118573ca46d31","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-17T14:30:06Z","title_canon_sha256":"403c21d96f5d82acd3e82fb7112262ae3a3835bec44594c0361b6136b8dd0510"},"schema_version":"1.0","source":{"id":"2506.14574","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.14574","created_at":"2026-07-05T11:22:59Z"},{"alias_kind":"arxiv_version","alias_value":"2506.14574v1","created_at":"2026-07-05T11:22:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.14574","created_at":"2026-07-05T11:22:59Z"},{"alias_kind":"pith_short_12","alias_value":"L5HEIOCSO3XS","created_at":"2026-07-05T11:22:59Z"},{"alias_kind":"pith_short_16","alias_value":"L5HEIOCSO3XSXF4F","created_at":"2026-07-05T11:22:59Z"},{"alias_kind":"pith_short_8","alias_value":"L5HEIOCS","created_at":"2026-07-05T11:22:59Z"}],"graph_snapshots":[{"event_id":"sha256:cbba1be37f3fed30daa186f67897daa642411628319e6c764e3b642a31390f66","target":"graph","created_at":"2026-07-05T11:22:59Z","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/2506.14574/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in reinforcement learning from human feedback have shown that utilizing fine-grained token-level reward models can substantially enhance the performance of Proximal Policy Optimization (PPO) in aligning large language models. However, it is challenging to leverage such token-level reward as guidance for Direct Preference Optimization (DPO), since DPO is formulated as a sequence-level bandit problem. To address this challenge, this work decomposes the sequence-level PPO into a sequence of token-level proximal policy optimization problems and then frames the problem of token-","authors_text":"Bei Yu, Hengshuang Zhao, Jiaya Jia, Mingkang Zhu, Xi Chen, Zhongdao Wang","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-17T14:30:06Z","title":"TGDPO: Harnessing Token-Level Reward Guidance for Enhancing Direct Preference Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.14574","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:d624202695e6c92bfed658d0daef798a9410eb2cfaff7f78b8cb4ca2095a2540","target":"record","created_at":"2026-07-05T11:22:59Z","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":"0814c105be5c38a9e76396cc46d681d73185561b2524c122fa2118573ca46d31","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-17T14:30:06Z","title_canon_sha256":"403c21d96f5d82acd3e82fb7112262ae3a3835bec44594c0361b6136b8dd0510"},"schema_version":"1.0","source":{"id":"2506.14574","kind":"arxiv","version":1}},"canonical_sha256":"5f4e44385276ef2b9785672612dc44cb6ee6e9bb6311dcafe0a9c667ad0f066f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5f4e44385276ef2b9785672612dc44cb6ee6e9bb6311dcafe0a9c667ad0f066f","first_computed_at":"2026-07-05T11:22:59.074047Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:22:59.074047Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Hxa45njDm0x0vOdaeMtpnKRhXbuZS+axVOXV0tlXs02Mqk9jvhRt7W2KKHrMKoP022As6GDqyM/32VsI6rz1BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:22:59.074554Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.14574","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d624202695e6c92bfed658d0daef798a9410eb2cfaff7f78b8cb4ca2095a2540","sha256:cbba1be37f3fed30daa186f67897daa642411628319e6c764e3b642a31390f66"],"state_sha256":"19e091aa067c34b6f9ede59cb4ce4d5f40d8fe7fe3dc1aa3e24dc2c5c45ee5d6"}