{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:2JI7OU7GKMHUVEERPP5I5JZ42Q","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":"8ef7eed017134e734d1977df226d7a57d53f1fdd628c324107e5b6b6fb3282af","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-23T18:01:11Z","title_canon_sha256":"9c377af5887133d4f53584a1bf438fa2fdd28f198a7bb8c6b8bce00ceb0f324c"},"schema_version":"1.0","source":{"id":"2405.14953","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.14953","created_at":"2026-07-05T10:50:42Z"},{"alias_kind":"arxiv_version","alias_value":"2405.14953v5","created_at":"2026-07-05T10:50:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14953","created_at":"2026-07-05T10:50:42Z"},{"alias_kind":"pith_short_12","alias_value":"2JI7OU7GKMHU","created_at":"2026-07-05T10:50:42Z"},{"alias_kind":"pith_short_16","alias_value":"2JI7OU7GKMHUVEER","created_at":"2026-07-05T10:50:42Z"},{"alias_kind":"pith_short_8","alias_value":"2JI7OU7G","created_at":"2026-07-05T10:50:42Z"}],"graph_snapshots":[{"event_id":"sha256:bccbc2581a879aebc63efd9b62a8eac1877ddfac7f4c056611dd04f2df7b276a","target":"graph","created_at":"2026-07-05T10:50:42Z","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/2405.14953/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Direct Preference Optimization (DPO) has recently emerged as a popular approach to improve reinforcement learning with human feedback (RLHF), leading to better techniques to fine-tune large language models (LLM). A weakness of DPO, however, lies in its lack of capability to characterize the diversity of human preferences. Inspired by Mallows' theory of preference ranking, we develop in this paper a new approach, the MallowsPO. A distinct feature of this approach is a dispersion index, which reflects the dispersion of human preference to prompts. We show that existing DPO models can be reduced ","authors_text":"David Yao, Hanyang Zhao, Haoxian Chen, Henry Lam, Wenpin Tang","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-23T18:01:11Z","title":"MallowsPO: Fine-Tune Your LLM with Preference Dispersions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14953","kind":"arxiv","version":5},"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":"sha256:4319493c8930a51e13ceb5a2e7e8439711139fa14a7da94d1772b10423262005","target":"record","created_at":"2026-07-05T10:50:42Z","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":"8ef7eed017134e734d1977df226d7a57d53f1fdd628c324107e5b6b6fb3282af","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-23T18:01:11Z","title_canon_sha256":"9c377af5887133d4f53584a1bf438fa2fdd28f198a7bb8c6b8bce00ceb0f324c"},"schema_version":"1.0","source":{"id":"2405.14953","kind":"arxiv","version":5}},"canonical_sha256":"d251f753e6530f4a90917bfa8ea73cd41759b9c5d26e700f5f9f90dd4f0bdc09","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d251f753e6530f4a90917bfa8ea73cd41759b9c5d26e700f5f9f90dd4f0bdc09","first_computed_at":"2026-07-05T10:50:42.208842Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:50:42.208842Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Vc7bLxAleiPjBTkO/ETeNUdygXTf8xFSjqy0iDcCdnrAFOu2whesm2T0HCFE7V9SQtm+2Jv0awglODEV2eGxDg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:50:42.209376Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.14953","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4319493c8930a51e13ceb5a2e7e8439711139fa14a7da94d1772b10423262005","sha256:bccbc2581a879aebc63efd9b62a8eac1877ddfac7f4c056611dd04f2df7b276a"],"state_sha256":"3578215eb3e96f5ec991e346414fd458ed572c3f6db3969f5b94261104cefb04"}