{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:DCQ36XWPSYQTLSWELKWV5IWTXU","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":"6cd2d764d97958f3cc77aaf687a4f65dbd6d7a0bafdfe5f8a2c1b2bc9a8ffc3d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-05T17:35:26Z","title_canon_sha256":"08f0f76097497bd7da786a924b5779b7bc5e83daf20d1e920839e538b902f2b2"},"schema_version":"1.0","source":{"id":"2310.03708","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.03708","created_at":"2026-07-05T08:56:14Z"},{"alias_kind":"arxiv_version","alias_value":"2310.03708v4","created_at":"2026-07-05T08:56:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.03708","created_at":"2026-07-05T08:56:14Z"},{"alias_kind":"pith_short_12","alias_value":"DCQ36XWPSYQT","created_at":"2026-07-05T08:56:14Z"},{"alias_kind":"pith_short_16","alias_value":"DCQ36XWPSYQTLSWE","created_at":"2026-07-05T08:56:14Z"},{"alias_kind":"pith_short_8","alias_value":"DCQ36XWP","created_at":"2026-07-05T08:56:14Z"}],"graph_snapshots":[{"event_id":"sha256:6a7fe0901c866054fa9c5fa02710a2d5b58cb659f834e3c90b4917ec4b045956","target":"graph","created_at":"2026-07-05T08:56:14Z","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/2310.03708/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A single language model, even when aligned with labelers through reinforcement learning from human feedback (RLHF), may not suit all human preferences. Recent approaches therefore prefer customization, gathering multi-dimensional feedback, and creating distinct reward models for each dimension. Different language models are then optimized for various preferences using multi-objective RLHF (MORLHF) with varying reward weights. However, RL fine-tuning is unstable and resource-heavy, especially with diverse and usually conflicting objectives. In this paper, we present Multi-Objective Direct Prefe","authors_text":"Chao Yang, Jie Liu, Jing Shao, Wanli Ouyang, Xiangyu Yue, Yu Qiao, Zhanhui Zhou","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-05T17:35:26Z","title":"Beyond One-Preference-Fits-All Alignment: Multi-Objective Direct Preference Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.03708","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:e96462ba5fc4e080e09a127a4281c248e3c974793b9e11524ba6d14a67302082","target":"record","created_at":"2026-07-05T08:56:14Z","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":"6cd2d764d97958f3cc77aaf687a4f65dbd6d7a0bafdfe5f8a2c1b2bc9a8ffc3d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-05T17:35:26Z","title_canon_sha256":"08f0f76097497bd7da786a924b5779b7bc5e83daf20d1e920839e538b902f2b2"},"schema_version":"1.0","source":{"id":"2310.03708","kind":"arxiv","version":4}},"canonical_sha256":"18a1bf5ecf962135cac45aad5ea2d3bd0dd5cb0b42d327bd4ae5c91cdb420473","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"18a1bf5ecf962135cac45aad5ea2d3bd0dd5cb0b42d327bd4ae5c91cdb420473","first_computed_at":"2026-07-05T08:56:14.364106Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:56:14.364106Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nrOJYhgDLe/TQIYnnEIVxVxt46o7m8dX60fLugDQXROYN1o2zflxRR0aXjiE+UBUyIWsnmv+wV7TRf3K4vw9Cg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:56:14.364515Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.03708","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e96462ba5fc4e080e09a127a4281c248e3c974793b9e11524ba6d14a67302082","sha256:6a7fe0901c866054fa9c5fa02710a2d5b58cb659f834e3c90b4917ec4b045956"],"state_sha256":"5b0bda4791d02688dfb4e7a0628aa61b4632561ac9120e1fc945ef19388f58f2"}