{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:Y4Y7GCJDRGVWASRXYHYQJGGOEA","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":"3f8d023bb8ef5138e7fe7bdc996e91293b1bd3a1cf4ccbe0ab62baec2dddd6fb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T09:18:02Z","title_canon_sha256":"f67714646b400e01626469cfd2185dedefb414c92bcb9fe40fd93d972cb0d077"},"schema_version":"1.0","source":{"id":"2502.12668","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.12668","created_at":"2026-07-05T10:15:59Z"},{"alias_kind":"arxiv_version","alias_value":"2502.12668v1","created_at":"2026-07-05T10:15:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12668","created_at":"2026-07-05T10:15:59Z"},{"alias_kind":"pith_short_12","alias_value":"Y4Y7GCJDRGVW","created_at":"2026-07-05T10:15:59Z"},{"alias_kind":"pith_short_16","alias_value":"Y4Y7GCJDRGVWASRX","created_at":"2026-07-05T10:15:59Z"},{"alias_kind":"pith_short_8","alias_value":"Y4Y7GCJD","created_at":"2026-07-05T10:15:59Z"}],"graph_snapshots":[{"event_id":"sha256:edc36f7a79de556665ea927df781a9bf5909b5d6801042f64aa42f6f78326b41","target":"graph","created_at":"2026-07-05T10:15:59Z","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/2502.12668/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Best-of-N (BoN) sampling with a reward model has been shown to be an effective strategy for aligning Large Language Models (LLMs) with human preferences at the time of decoding. BoN sampling is susceptible to a problem known as reward hacking. Since the reward model is an imperfect proxy for the true objective, an excessive focus on optimizing its value can lead to a compromise of its performance on the true objective. Previous work proposes Regularized BoN sampling (RBoN), a BoN sampling with regularization to the objective, and shows that it outperforms BoN sampling so that it mitigates rewa","authors_text":"Eiji Uchibe, Kaito Ariu, Kenshi Abe, Mitsuki Sakamoto, Tetsuro Morimura, Yuki Ichihara, Yuu Jinnai","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T09:18:02Z","title":"Evaluation of Best-of-N Sampling Strategies for Language Model Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12668","kind":"arxiv","version":1},"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":"sha256:3248f34d039b2fa23ca3908a4bcbd688e705f10be83e327eab8a171cee7cfdcc","target":"record","created_at":"2026-07-05T10:15:59Z","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":"3f8d023bb8ef5138e7fe7bdc996e91293b1bd3a1cf4ccbe0ab62baec2dddd6fb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T09:18:02Z","title_canon_sha256":"f67714646b400e01626469cfd2185dedefb414c92bcb9fe40fd93d972cb0d077"},"schema_version":"1.0","source":{"id":"2502.12668","kind":"arxiv","version":1}},"canonical_sha256":"c731f3092389ab604a37c1f10498ce203535a72b24a43a36267f561ce4c917a8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c731f3092389ab604a37c1f10498ce203535a72b24a43a36267f561ce4c917a8","first_computed_at":"2026-07-05T10:15:59.136363Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:15:59.136363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Q9nsuGMWnSOGPeaXOq3k9hv8EenwTwGzUJ0VxAzoDSd3eZjiFcHZovbwLil8MXCL8750VxMx/20mVfbEVD0PDg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:15:59.136778Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.12668","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3248f34d039b2fa23ca3908a4bcbd688e705f10be83e327eab8a171cee7cfdcc","sha256:edc36f7a79de556665ea927df781a9bf5909b5d6801042f64aa42f6f78326b41"],"state_sha256":"47620e8441ecd7a76e39e49099c4834ca6a8e7063bd192777ffdddc1a18d2d70"}