{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:W53EKNHYKC5XIA73ZVS74SNMT3","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":"231082eb49184e242b88fad3165bd23a7a1397ec621967d34cdfc4dddc40d2cf","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-22T16:24:43Z","title_canon_sha256":"8a7b9763e35d528c15aaf60781f85366b95fe54e5d5373ef01a9721b9d59a00b"},"schema_version":"1.0","source":{"id":"2401.12086","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.12086","created_at":"2026-07-05T09:25:38Z"},{"alias_kind":"arxiv_version","alias_value":"2401.12086v2","created_at":"2026-07-05T09:25:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.12086","created_at":"2026-07-05T09:25:38Z"},{"alias_kind":"pith_short_12","alias_value":"W53EKNHYKC5X","created_at":"2026-07-05T09:25:38Z"},{"alias_kind":"pith_short_16","alias_value":"W53EKNHYKC5XIA73","created_at":"2026-07-05T09:25:38Z"},{"alias_kind":"pith_short_8","alias_value":"W53EKNHY","created_at":"2026-07-05T09:25:38Z"}],"graph_snapshots":[{"event_id":"sha256:e7b6848f39cb805395a0e334f1e1a01013e933255e76f1932f2c03eb792a4399","target":"graph","created_at":"2026-07-05T09:25:38Z","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/2401.12086/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The success of reinforcement learning from human feedback (RLHF) in language model alignment is strongly dependent on the quality of the underlying reward model. In this paper, we present a novel approach to improve reward model quality by generating synthetic preference data, thereby augmenting the training dataset with on-policy, high-quality preference pairs. Motivated by the promising results of Best-of-N sampling strategies in language model training, we extend their application to reward model training. This results in a self-training strategy to generate preference pairs by selecting th","authors_text":"Aliaksei Severyn, Aliz\\'ee Pace, Eric Malmi, Jonathan Mallinson, Sebastian Krause","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-22T16:24:43Z","title":"West-of-N: Synthetic Preferences for Self-Improving Reward Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.12086","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":"sha256:768ae52d7dbc284437c733b2466dbafca054d5f41e1c346545f1c6c382706bf6","target":"record","created_at":"2026-07-05T09:25:38Z","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":"231082eb49184e242b88fad3165bd23a7a1397ec621967d34cdfc4dddc40d2cf","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-22T16:24:43Z","title_canon_sha256":"8a7b9763e35d528c15aaf60781f85366b95fe54e5d5373ef01a9721b9d59a00b"},"schema_version":"1.0","source":{"id":"2401.12086","kind":"arxiv","version":2}},"canonical_sha256":"b7764534f850bb7403fbcd65fe49ac9ec5a539966cc4f56801550ace3e4131d3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b7764534f850bb7403fbcd65fe49ac9ec5a539966cc4f56801550ace3e4131d3","first_computed_at":"2026-07-05T09:25:38.729894Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:25:38.729894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XE7L4Z4IL5XL1SFi/XBdKvbkLhJorR32iWUdg6yYXxJ/5kwmFzTmcUZyGEXR2UYSxSLaFzz6A4EzLopUqxIsAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:25:38.730411Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.12086","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:768ae52d7dbc284437c733b2466dbafca054d5f41e1c346545f1c6c382706bf6","sha256:e7b6848f39cb805395a0e334f1e1a01013e933255e76f1932f2c03eb792a4399"],"state_sha256":"2509d2dbc4548ba7663c066d8abb15034ca0cf8b4f9542aed17b41282f49c89f"}