{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UWZSZDCNE5ZLW4MBWDPH4PRQOG","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":"cf520f99905c65841559b4016df16e4c3a7d3c2e15254ab072c7d2322d90f2f9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-28T18:02:35Z","title_canon_sha256":"fba430344ee564341a8eeaaf413c784196d06b469a3b61aef11df8c72ee48635"},"schema_version":"1.0","source":{"id":"2504.20157","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.20157","created_at":"2026-07-05T11:04:21Z"},{"alias_kind":"arxiv_version","alias_value":"2504.20157v2","created_at":"2026-07-05T11:04:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.20157","created_at":"2026-07-05T11:04:21Z"},{"alias_kind":"pith_short_12","alias_value":"UWZSZDCNE5ZL","created_at":"2026-07-05T11:04:21Z"},{"alias_kind":"pith_short_16","alias_value":"UWZSZDCNE5ZLW4MB","created_at":"2026-07-05T11:04:21Z"},{"alias_kind":"pith_short_8","alias_value":"UWZSZDCN","created_at":"2026-07-05T11:04:21Z"}],"graph_snapshots":[{"event_id":"sha256:0cc4dd8a74f5595222d9a3add172c493c8b1a82822a9968bacb428f231dc6ece","target":"graph","created_at":"2026-07-05T11:04:21Z","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/2504.20157/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reward-based alignment methods for large language models (LLMs) face two key limitations: vulnerability to reward hacking, where models exploit flaws in the reward signal; and reliance on brittle, labor-intensive prompt engineering when LLMs are used as reward models. We introduce Meta Policy Optimization (MPO), a framework that addresses these challenges by integrating a meta-reward model that dynamically refines the reward model's prompt throughout training. In MPO, the meta-reward model monitors the evolving training context and continuously adjusts the reward model's prompt to maintain hig","authors_text":"Chanwoo Park, Dongyeop Kang, Suin Kim, Vipul Raheja, Zae Myung Kim","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-28T18:02:35Z","title":"Toward Evaluative Thinking: Meta Policy Optimization with Evolving Reward Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.20157","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:e523140ba8d9b6062c461441e6a76700298ef7013a45cf22340b77e462ac01b6","target":"record","created_at":"2026-07-05T11:04:21Z","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":"cf520f99905c65841559b4016df16e4c3a7d3c2e15254ab072c7d2322d90f2f9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-28T18:02:35Z","title_canon_sha256":"fba430344ee564341a8eeaaf413c784196d06b469a3b61aef11df8c72ee48635"},"schema_version":"1.0","source":{"id":"2504.20157","kind":"arxiv","version":2}},"canonical_sha256":"a5b32c8c4d2772bb7181b0de7e3e3071932ee580b045a821d8afba7b1e7f3da2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a5b32c8c4d2772bb7181b0de7e3e3071932ee580b045a821d8afba7b1e7f3da2","first_computed_at":"2026-07-05T11:04:21.895711Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:04:21.895711Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"M+SotEYWNLx8QdFzUk12FsYbkO/z1fhHywgHH93O0tHgq+bIfuNDiyBxQg9UwLcVAgFDb20gOAufmp/eSz5iCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:04:21.896232Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.20157","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e523140ba8d9b6062c461441e6a76700298ef7013a45cf22340b77e462ac01b6","sha256:0cc4dd8a74f5595222d9a3add172c493c8b1a82822a9968bacb428f231dc6ece"],"state_sha256":"1dd9208762705e804c64fedb269eedbc54d528b0995b27e5ca1b4a617b7f3b00"}