{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7PXHOR7SBD7UOEXXPARCUJ6D3P","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":"538c8b449132aac3b1280bd4b74cea985cda293875ae35a7a3eabad3367d3a1c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T16:53:08Z","title_canon_sha256":"7041f472aca480113a09e19aaeb68df5cdaec94cf6302d9f7c7817ea0df0d5d5"},"schema_version":"1.0","source":{"id":"2501.13011","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.13011","created_at":"2026-07-05T10:47:20Z"},{"alias_kind":"arxiv_version","alias_value":"2501.13011v2","created_at":"2026-07-05T10:47:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.13011","created_at":"2026-07-05T10:47:20Z"},{"alias_kind":"pith_short_12","alias_value":"7PXHOR7SBD7U","created_at":"2026-07-05T10:47:20Z"},{"alias_kind":"pith_short_16","alias_value":"7PXHOR7SBD7UOEXX","created_at":"2026-07-05T10:47:20Z"},{"alias_kind":"pith_short_8","alias_value":"7PXHOR7S","created_at":"2026-07-05T10:47:20Z"}],"graph_snapshots":[{"event_id":"sha256:c28aace90783577da485e34f63a7b5ab8f05a4d11ed9079ca82daee4d7567003","target":"graph","created_at":"2026-07-05T10:47:20Z","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/2501.13011/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Future advanced AI systems may learn sophisticated strategies through reinforcement learning (RL) that humans cannot understand well enough to safely evaluate. We propose a training method which avoids agents learning undesired multi-step plans that receive high reward (multi-step \"reward hacks\") even if humans are not able to detect that the behaviour is undesired. The method, Myopic Optimization with Non-myopic Approval (MONA), works by combining short-sighted optimization with far-sighted reward. We demonstrate that MONA can prevent multi-step reward hacking that ordinary RL causes, even wi","authors_text":"Caleb Biddulph, David Elson, David Lindner, Ian Goodfellow, Rohin Shah, Sebastian Farquhar, Vikrant Varma","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T16:53:08Z","title":"MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.13011","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":"sha256:25f676c9090a3c272220ff75e5a0af7e4f4947ff836cdeb271879d5f542c067b","target":"record","created_at":"2026-07-05T10:47:20Z","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":"538c8b449132aac3b1280bd4b74cea985cda293875ae35a7a3eabad3367d3a1c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-22T16:53:08Z","title_canon_sha256":"7041f472aca480113a09e19aaeb68df5cdaec94cf6302d9f7c7817ea0df0d5d5"},"schema_version":"1.0","source":{"id":"2501.13011","kind":"arxiv","version":2}},"canonical_sha256":"fbee7747f208ff4712f778222a27c3dbe0264c5a9459467b17768f0e7636c136","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fbee7747f208ff4712f778222a27c3dbe0264c5a9459467b17768f0e7636c136","first_computed_at":"2026-07-05T10:47:20.874349Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:47:20.874349Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AKU9W55/goHUrXRTTaGRxChxnqYNXhTnDnCqhjas9raVDOmKwL+MpPlT744xgYK1Utkw62rh7eUjQ+k3rreYDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:47:20.874811Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.13011","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:25f676c9090a3c272220ff75e5a0af7e4f4947ff836cdeb271879d5f542c067b","sha256:c28aace90783577da485e34f63a7b5ab8f05a4d11ed9079ca82daee4d7567003"],"state_sha256":"c7bbce11c05ec143cbfb94a4768d5180329d40fa88a61eb93d27898e2d07976f"}