{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SFR3YRL5V2FNY65MMC5ABUAPNI","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":"28f8a251203274ae9ab9566b663177037c4252f701a6cd2d3b93074556339454","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-10-03T04:22:55Z","title_canon_sha256":"e757f151e6648d354fc8c7206c7a60b2c777afc4089849b24e3cff089db2002f"},"schema_version":"1.0","source":{"id":"2410.02197","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.02197","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"arxiv_version","alias_value":"2410.02197v3","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.02197","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"pith_short_12","alias_value":"SFR3YRL5V2FN","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"pith_short_16","alias_value":"SFR3YRL5V2FNY65M","created_at":"2026-07-05T11:19:49Z"},{"alias_kind":"pith_short_8","alias_value":"SFR3YRL5","created_at":"2026-07-05T11:19:49Z"}],"graph_snapshots":[{"event_id":"sha256:14281f6138b0ba0b79f45d2c81779eec8d5300b26706dff8bdeddf8f74fdc797","target":"graph","created_at":"2026-07-05T11:19:49Z","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/2410.02197/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modeling human preferences is crucial for aligning foundation models with human values. Traditional reward modeling methods, such as the Bradley-Terry (BT) reward model, fall short in expressiveness, particularly in addressing intransitive preferences. In this paper, we introduce preference embedding, an approach that embeds responses into a latent space to capture intricate preference structures efficiently, achieving linear query complexity. Additionally, we propose preference score-based General Preference Optimization (GPO), which generalizes reward-based reinforcement learning from human ","authors_text":"Ge Zhang, Kangping Xu, Quanquan Gu, Yifan Zhang, Yue Wu","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-10-03T04:22:55Z","title":"Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.02197","kind":"arxiv","version":3},"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":"sha256:77db54f380643bd31586cce696ecd3c01725eefed16a208b159c12dfdbe4231b","target":"record","created_at":"2026-07-05T11:19:49Z","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":"28f8a251203274ae9ab9566b663177037c4252f701a6cd2d3b93074556339454","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-10-03T04:22:55Z","title_canon_sha256":"e757f151e6648d354fc8c7206c7a60b2c777afc4089849b24e3cff089db2002f"},"schema_version":"1.0","source":{"id":"2410.02197","kind":"arxiv","version":3}},"canonical_sha256":"9163bc457dae8adc7bac60ba00d00f6a0cecef43fec217056c7be5f1c7ebc333","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9163bc457dae8adc7bac60ba00d00f6a0cecef43fec217056c7be5f1c7ebc333","first_computed_at":"2026-07-05T11:19:49.582616Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:49.582616Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XMQCNJ1b+arm1g325ThZb1ahgnAPMU9kTilCrY26x1sK6ZxthCJcBiuhhh1VYlUoU0P/wEKBMEce8Xg0RsepAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:49.583073Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.02197","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:77db54f380643bd31586cce696ecd3c01725eefed16a208b159c12dfdbe4231b","sha256:14281f6138b0ba0b79f45d2c81779eec8d5300b26706dff8bdeddf8f74fdc797"],"state_sha256":"11b51092bab2349349faf569f449e1f4253168db74669c8d25cc976636fb2f40"}