{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:PATWTZHH76AL2B6SUZKU6OI3WE","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":"652324c984b05f7e7632eb87cc7fa48b4266f2a806dea534d0f80f855f4c4de0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-07-04T23:26:56Z","title_canon_sha256":"93dc99ecb9bc46967c239c23b5635822f9b96af17429f1ee8d771084b0e6cb46"},"schema_version":"1.0","source":{"id":"2407.04185","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.04185","created_at":"2026-07-05T09:58:22Z"},{"alias_kind":"arxiv_version","alias_value":"2407.04185v4","created_at":"2026-07-05T09:58:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.04185","created_at":"2026-07-05T09:58:22Z"},{"alias_kind":"pith_short_12","alias_value":"PATWTZHH76AL","created_at":"2026-07-05T09:58:22Z"},{"alias_kind":"pith_short_16","alias_value":"PATWTZHH76AL2B6S","created_at":"2026-07-05T09:58:22Z"},{"alias_kind":"pith_short_8","alias_value":"PATWTZHH","created_at":"2026-07-05T09:58:22Z"}],"graph_snapshots":[{"event_id":"sha256:bae89334d9dd23ec896c0b8e4d2159481a9440c047f80964743b7fa3dfb972b9","target":"graph","created_at":"2026-07-05T09:58:22Z","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/2407.04185/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The reward model has become increasingly important in alignment, assessment, and data construction for large language models (LLMs). Most existing researchers focus on enhancing reward models through data improvements, following the conventional training framework for reward models that directly optimizes the predicted rewards. In this paper, we propose a hybrid alignment framework HaF-RM for reward model training by introducing an additional constraint on token-level policy probabilities in addition to the reward score. It can simultaneously supervise the internal preference model at the toke","authors_text":"Shengbin Yue, Shujun Liu, Siyuan Wang, Xiaoyu Shen, Xuanjing Huang, Yuhang Lai, Zengfeng Huang, Zhongyu Wei","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-07-04T23:26:56Z","title":"HAF-RM: A Hybrid Alignment Framework for Reward Model Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.04185","kind":"arxiv","version":4},"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:f781a9e7679aa1b317d4d9cd19e58de6997d8e46b86eda8880e30c6424e4e360","target":"record","created_at":"2026-07-05T09:58:22Z","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":"652324c984b05f7e7632eb87cc7fa48b4266f2a806dea534d0f80f855f4c4de0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-07-04T23:26:56Z","title_canon_sha256":"93dc99ecb9bc46967c239c23b5635822f9b96af17429f1ee8d771084b0e6cb46"},"schema_version":"1.0","source":{"id":"2407.04185","kind":"arxiv","version":4}},"canonical_sha256":"782769e4e7ff80bd07d2a6554f391bb12bf4357657481cf3911919e471a34f07","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"782769e4e7ff80bd07d2a6554f391bb12bf4357657481cf3911919e471a34f07","first_computed_at":"2026-07-05T09:58:22.344848Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:58:22.344848Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"q5JhrdzZqUj8DanRC6+wIuAO2t4Q6gUKBvGQf/XpcJZ9ML0g6nVgzKmUNeRCfvvT0Xmno5Eq/Oax1wtZX7gqDA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:58:22.345259Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.04185","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f781a9e7679aa1b317d4d9cd19e58de6997d8e46b86eda8880e30c6424e4e360","sha256:bae89334d9dd23ec896c0b8e4d2159481a9440c047f80964743b7fa3dfb972b9"],"state_sha256":"ab525b3291c635d5ccf40d6d77027498b4a13790b81fec98d38f6c0700fdd380"}