{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3DCJBUBJB4I2KDPYRGZ7IJTBWF","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":"ead0103d1306794ee5a3f335a09d4ad04f832a8cee0020327d7f6d156b77a615","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-16T00:59:19Z","title_canon_sha256":"6611c777c973fb3c57f4fae5782128bd88f159641dfaaded1df7c084dbfa4e11"},"schema_version":"1.0","source":{"id":"2410.12138","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.12138","created_at":"2026-07-05T10:39:21Z"},{"alias_kind":"arxiv_version","alias_value":"2410.12138v2","created_at":"2026-07-05T10:39:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.12138","created_at":"2026-07-05T10:39:21Z"},{"alias_kind":"pith_short_12","alias_value":"3DCJBUBJB4I2","created_at":"2026-07-05T10:39:21Z"},{"alias_kind":"pith_short_16","alias_value":"3DCJBUBJB4I2KDPY","created_at":"2026-07-05T10:39:21Z"},{"alias_kind":"pith_short_8","alias_value":"3DCJBUBJ","created_at":"2026-07-05T10:39:21Z"}],"graph_snapshots":[{"event_id":"sha256:ab282ad9c5a30c2877457e961e1568e4cc2c394af06d517d8b8bb2a511aa327f","target":"graph","created_at":"2026-07-05T10:39:21Z","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/2410.12138/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in generative models, particularly large language models (LLMs) and diffusion models, have been driven by extensive pretraining on large datasets followed by post-training. However, current post-training methods such as reinforcement learning from human feedback (RLHF) and direct alignment from preference methods (DAP) primarily utilize single-sample comparisons. These approaches often fail to capture critical characteristics such as generative diversity and bias, which are more accurately assessed through multiple samples. To address these limitations, we introduce a novel","authors_text":"Chaoqi Wang, Chen Zhu, Hao Ma, Karthik Abinav Sankararaman, Madian Khabsa, Michal Valko, Sinong Wang, Xuefei Cao, Yuxin Chen, Zhaorun Chen, Zhuokai Zhao","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-16T00:59:19Z","title":"Preference Optimization with Multi-Sample Comparisons"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.12138","kind":"arxiv","version":2},"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:bde988c1242e3338ba9dc3cd45db3d72defeb2665e1ac091dbfb4df57f946945","target":"record","created_at":"2026-07-05T10:39:21Z","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":"ead0103d1306794ee5a3f335a09d4ad04f832a8cee0020327d7f6d156b77a615","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-16T00:59:19Z","title_canon_sha256":"6611c777c973fb3c57f4fae5782128bd88f159641dfaaded1df7c084dbfa4e11"},"schema_version":"1.0","source":{"id":"2410.12138","kind":"arxiv","version":2}},"canonical_sha256":"d8c490d0290f11a50df889b3f42661b1694e827e0dd80df08111525c56b85d79","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d8c490d0290f11a50df889b3f42661b1694e827e0dd80df08111525c56b85d79","first_computed_at":"2026-07-05T10:39:21.792494Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:39:21.792494Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KqcvFfOhpPdP6NfUGNUublcbNPrFu5BqTbmZKWraOw7J7CKXRFsyU3QvZLvZjK5f6na6tMEJgOLrQuuyJK0VAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:39:21.792988Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.12138","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bde988c1242e3338ba9dc3cd45db3d72defeb2665e1ac091dbfb4df57f946945","sha256:ab282ad9c5a30c2877457e961e1568e4cc2c394af06d517d8b8bb2a511aa327f"],"state_sha256":"88f0645fa2c661fc9aaa72e2b90eaaeb1e8006645edf32ece7aeb1ca067d4d89"}