{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:DJRWIYNUUJJ6H4DOUZMUHEBJWB","short_pith_number":"pith:DJRWIYNU","canonical_record":{"source":{"id":"2406.02900","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-05T03:41:37Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"e5d73989fe210f83270294d1a61ea10742618f1b9b9c8c08fdeea49a005f2b29","abstract_canon_sha256":"fb051add89d3b124e94d0c64c25b9bfb4fdc1eeb222ffa2b685d5db35a82e67e"},"schema_version":"1.0"},"canonical_sha256":"1a636461b4a253e3f06ea659439029b05f510d6d27c044a766345258c75a8644","source":{"kind":"arxiv","id":"2406.02900","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.02900","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"arxiv_version","alias_value":"2406.02900v2","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.02900","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"pith_short_12","alias_value":"DJRWIYNUUJJ6","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"pith_short_16","alias_value":"DJRWIYNUUJJ6H4DO","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"pith_short_8","alias_value":"DJRWIYNU","created_at":"2026-07-05T09:31:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:DJRWIYNUUJJ6H4DOUZMUHEBJWB","target":"record","payload":{"canonical_record":{"source":{"id":"2406.02900","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-05T03:41:37Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"e5d73989fe210f83270294d1a61ea10742618f1b9b9c8c08fdeea49a005f2b29","abstract_canon_sha256":"fb051add89d3b124e94d0c64c25b9bfb4fdc1eeb222ffa2b685d5db35a82e67e"},"schema_version":"1.0"},"canonical_sha256":"1a636461b4a253e3f06ea659439029b05f510d6d27c044a766345258c75a8644","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:31:03.720365Z","signature_b64":"fG/aC6ZUIrfggUgMkoWHI4xDOoUAUcuIv8cvpml7zhe8zyPrAS55l0hOEa8b0y79Cbo0WSHAlPJxxLn66TB5AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a636461b4a253e3f06ea659439029b05f510d6d27c044a766345258c75a8644","last_reissued_at":"2026-07-05T09:31:03.719850Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:31:03.719850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.02900","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:31:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"55I8kT8R8VEXAoiEeiRWgK/D71YqpzgSCBOGm6VVQWRo0EJgH6d9vJ4lPuUu190+AyPEev+CpI0CleV040HDAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T11:08:55.926268Z"},"content_sha256":"2d490c2224ac3138f5032df509c6ac3e1140310c660b77cc355266ad4bf37368","schema_version":"1.0","event_id":"sha256:2d490c2224ac3138f5032df509c6ac3e1140310c660b77cc355266ad4bf37368"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:DJRWIYNUUJJ6H4DOUZMUHEBJWB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Scaling Laws for Reward Model Overoptimization in Direct Alignment Algorithms","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bradley Knox, Chelsea Finn, Harshit Sikchi, Joey Hejna, Rafael Rafailov, Ryan Park, Scott Niekum, Yaswanth Chittepu","submitted_at":"2024-06-05T03:41:37Z","abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) has been crucial to the recent success of Large Language Models (LLMs), however, it is often a complex and brittle process. In the classical RLHF framework, a reward model is first trained to represent human preferences, which is in turn used by an online reinforcement learning (RL) algorithm to optimize the LLM. A prominent issue with such methods is reward over-optimization or reward hacking, where performance as measured by the learned proxy reward model increases, but true quality plateaus or even deteriorates. Direct Alignment Algorithms ("},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.02900","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2406.02900/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:31:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IVhiGC/25M8EvbD1s0z7H2/KXRRPhos1D+97xh3198is4Rat7AnfTxFr2Q8hpYJyPdVlst/5mdF7AMp5rua/Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T11:08:55.926783Z"},"content_sha256":"162886539ef9f55ebbb457c54acdbaed53d29df630bd0debe10d467cd4bb082c","schema_version":"1.0","event_id":"sha256:162886539ef9f55ebbb457c54acdbaed53d29df630bd0debe10d467cd4bb082c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DJRWIYNUUJJ6H4DOUZMUHEBJWB/bundle.json","state_url":"https://pith.science/pith/DJRWIYNUUJJ6H4DOUZMUHEBJWB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DJRWIYNUUJJ6H4DOUZMUHEBJWB/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-18T11:08:55Z","links":{"resolver":"https://pith.science/pith/DJRWIYNUUJJ6H4DOUZMUHEBJWB","bundle":"https://pith.science/pith/DJRWIYNUUJJ6H4DOUZMUHEBJWB/bundle.json","state":"https://pith.science/pith/DJRWIYNUUJJ6H4DOUZMUHEBJWB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DJRWIYNUUJJ6H4DOUZMUHEBJWB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DJRWIYNUUJJ6H4DOUZMUHEBJWB","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":"fb051add89d3b124e94d0c64c25b9bfb4fdc1eeb222ffa2b685d5db35a82e67e","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-05T03:41:37Z","title_canon_sha256":"e5d73989fe210f83270294d1a61ea10742618f1b9b9c8c08fdeea49a005f2b29"},"schema_version":"1.0","source":{"id":"2406.02900","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.02900","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"arxiv_version","alias_value":"2406.02900v2","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.02900","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"pith_short_12","alias_value":"DJRWIYNUUJJ6","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"pith_short_16","alias_value":"DJRWIYNUUJJ6H4DO","created_at":"2026-07-05T09:31:03Z"},{"alias_kind":"pith_short_8","alias_value":"DJRWIYNU","created_at":"2026-07-05T09:31:03Z"}],"graph_snapshots":[{"event_id":"sha256:162886539ef9f55ebbb457c54acdbaed53d29df630bd0debe10d467cd4bb082c","target":"graph","created_at":"2026-07-05T09:31:03Z","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/2406.02900/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) has been crucial to the recent success of Large Language Models (LLMs), however, it is often a complex and brittle process. In the classical RLHF framework, a reward model is first trained to represent human preferences, which is in turn used by an online reinforcement learning (RL) algorithm to optimize the LLM. A prominent issue with such methods is reward over-optimization or reward hacking, where performance as measured by the learned proxy reward model increases, but true quality plateaus or even deteriorates. Direct Alignment Algorithms (","authors_text":"Bradley Knox, Chelsea Finn, Harshit Sikchi, Joey Hejna, Rafael Rafailov, Ryan Park, Scott Niekum, Yaswanth Chittepu","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-05T03:41:37Z","title":"Scaling Laws for Reward Model Overoptimization in Direct Alignment Algorithms"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.02900","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":"sha256:2d490c2224ac3138f5032df509c6ac3e1140310c660b77cc355266ad4bf37368","target":"record","created_at":"2026-07-05T09:31:03Z","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":"fb051add89d3b124e94d0c64c25b9bfb4fdc1eeb222ffa2b685d5db35a82e67e","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-05T03:41:37Z","title_canon_sha256":"e5d73989fe210f83270294d1a61ea10742618f1b9b9c8c08fdeea49a005f2b29"},"schema_version":"1.0","source":{"id":"2406.02900","kind":"arxiv","version":2}},"canonical_sha256":"1a636461b4a253e3f06ea659439029b05f510d6d27c044a766345258c75a8644","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1a636461b4a253e3f06ea659439029b05f510d6d27c044a766345258c75a8644","first_computed_at":"2026-07-05T09:31:03.719850Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:31:03.719850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fG/aC6ZUIrfggUgMkoWHI4xDOoUAUcuIv8cvpml7zhe8zyPrAS55l0hOEa8b0y79Cbo0WSHAlPJxxLn66TB5AA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:31:03.720365Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.02900","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2d490c2224ac3138f5032df509c6ac3e1140310c660b77cc355266ad4bf37368","sha256:162886539ef9f55ebbb457c54acdbaed53d29df630bd0debe10d467cd4bb082c"],"state_sha256":"e2af260c7ba240a820254f60961e3dacede651839655ac9d10c21f828a897f02"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QUzAp4ygebOloqm1jqGpEjq93DX9gCVbTxYCISCR+o2J7Nujifjf7OZRtLfStmMSpQPyBaXCCn4ZhtWJIadmBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T11:08:55.930683Z","bundle_sha256":"27adaf3d240b2c05c8c6dbd1236a94209b1592b2e7a8e75120f974b6c279037c"}}