{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:23IZB7K2H2STHS5OGU3SX45Y43","short_pith_number":"pith:23IZB7K2","canonical_record":{"source":{"id":"2307.15217","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-07-27T22:29:25Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"f64b8e348d873883aaed9189c75a52a859f0648b3f54a9627b55a100e94a8e16","abstract_canon_sha256":"b2d5a1ffd2f126b3687464bdd54878590cc00c30071a32d643bc8ba98db2121f"},"schema_version":"1.0"},"canonical_sha256":"d6d190fd5a3ea533cbae35372bf3b8e6ddf630437eb272f01b14c5a437f010e0","source":{"kind":"arxiv","id":"2307.15217","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.15217","created_at":"2026-05-17T23:39:21Z"},{"alias_kind":"arxiv_version","alias_value":"2307.15217v2","created_at":"2026-05-17T23:39:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.15217","created_at":"2026-05-17T23:39:21Z"},{"alias_kind":"pith_short_12","alias_value":"23IZB7K2H2ST","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"23IZB7K2H2STHS5O","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"23IZB7K2","created_at":"2026-05-18T12:33:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:23IZB7K2H2STHS5OGU3SX45Y43","target":"record","payload":{"canonical_record":{"source":{"id":"2307.15217","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-07-27T22:29:25Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"f64b8e348d873883aaed9189c75a52a859f0648b3f54a9627b55a100e94a8e16","abstract_canon_sha256":"b2d5a1ffd2f126b3687464bdd54878590cc00c30071a32d643bc8ba98db2121f"},"schema_version":"1.0"},"canonical_sha256":"d6d190fd5a3ea533cbae35372bf3b8e6ddf630437eb272f01b14c5a437f010e0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:21.419160Z","signature_b64":"IgWYAzu8E11YJKSnEBOZRO42bFde2+JBCzbLVCW8GfzIcKwkkklGwB+qS0rz/wj0+t86Er3LpC7yetHSWlyQCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d6d190fd5a3ea533cbae35372bf3b8e6ddf630437eb272f01b14c5a437f010e0","last_reissued_at":"2026-05-17T23:39:21.418541Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:21.418541Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2307.15217","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":"2026-05-17T23:39:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"don3JszoWI6iVExaCcx4gUc6SkF7YamTY1YgAkF4t2FRBLlH5GcbfDGauevqjsiUOnFBbUBpKBLH1rG4gKqbAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T16:56:19.592380Z"},"content_sha256":"b62ab5e7af6944a12434dc205184fc460cdc7cfcadabd3ea5c37b08308edc198","schema_version":"1.0","event_id":"sha256:b62ab5e7af6944a12434dc205184fc460cdc7cfcadabd3ea5c37b08308edc198"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:23IZB7K2H2STHS5OGU3SX45Y43","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"RLHF, the dominant method for aligning large language models with human goals, carries fundamental limitations that incremental fixes cannot fully resolve.","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Anand Siththaranjan, Anca Dragan, Andi Peng, Charbel-Rapha\\\"el Segerie, Claudia Shi, David Krueger, David Lindner, Dmitrii Krasheninnikov, Dorsa Sadigh, Dylan Hadfield-Menell, Erdem B{\\i}y{\\i}k, Eric J. Michaud, Jacob Pfau, Javier Rando, J\\'er\\'emy Scheurer, Lauro Langosco, Max Nadeau, Mehul Damani, Micah Carroll, Pedro Freire, Peter Hase, Phillip Christoffersen, Rachel Freedman, Samuel Marks, Stephen Casper, Stewart Slocum, Thomas Krendl Gilbert, Tomasz Korbak, Tony Wang, Usman Anwar, Xander Davies, Xin Chen","submitted_at":"2023-07-27T22:29:25Z","abstract_excerpt":"Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there has been relatively little public work systematizing its flaws. In this paper, we (1) survey open problems and fundamental limitations of RLHF and related methods; (2) overview techniques to understand, improve, and complement RLHF in practice; and (3) propose auditing and disclosure standards to improve societal oversight of RLHF systems. Our work"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"RLHF has emerged as the central method used to finetune state-of-the-art large language models but has fundamental limitations, emphasizing the importance of a multi-faceted approach to the development of safer AI systems.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the identified open problems represent fundamental limitations of RLHF rather than challenges that can be resolved through incremental improvements or better implementation.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"RLHF has significant open problems and fundamental limitations that require a multi-faceted approach for safer AI development.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"RLHF, the dominant method for aligning large language models with human goals, carries fundamental limitations that incremental fixes cannot fully resolve.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"006f1672e3aad5d3409d2fe0f53106294ab14aa905c8916cb46d1259040d51e5"},"source":{"id":"2307.15217","kind":"arxiv","version":2},"verdict":{"id":"1928dcfc-c59b-4bc6-a627-8307e124f306","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T22:40:12.619170Z","strongest_claim":"RLHF has emerged as the central method used to finetune state-of-the-art large language models but has fundamental limitations, emphasizing the importance of a multi-faceted approach to the development of safer AI systems.","one_line_summary":"RLHF has significant open problems and fundamental limitations that require a multi-faceted approach for safer AI development.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the identified open problems represent fundamental limitations of RLHF rather than challenges that can be resolved through incremental improvements or better implementation.","pith_extraction_headline":"RLHF, the dominant method for aligning large language models with human goals, carries fundamental limitations that incremental fixes cannot fully resolve."},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"5803d0de03d454fe4fec6249f84c86bea3f67e278651388713b656953744c67d"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"1928dcfc-c59b-4bc6-a627-8307e124f306"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+7lsJ54krBUBSLP/HUq0WOSJbRaFppR7USy5Rn994/19FvCnION95WzuiOi26VvgmJN7eMy4HNhpD9wroV9rCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T16:56:19.593256Z"},"content_sha256":"4fde2cec5ebbdf1d6924a489b435f99e8ffc28db3a1b8cad8e2a461ba984061a","schema_version":"1.0","event_id":"sha256:4fde2cec5ebbdf1d6924a489b435f99e8ffc28db3a1b8cad8e2a461ba984061a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/23IZB7K2H2STHS5OGU3SX45Y43/bundle.json","state_url":"https://pith.science/pith/23IZB7K2H2STHS5OGU3SX45Y43/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/23IZB7K2H2STHS5OGU3SX45Y43/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":"2026-05-25T16:56:19Z","links":{"resolver":"https://pith.science/pith/23IZB7K2H2STHS5OGU3SX45Y43","bundle":"https://pith.science/pith/23IZB7K2H2STHS5OGU3SX45Y43/bundle.json","state":"https://pith.science/pith/23IZB7K2H2STHS5OGU3SX45Y43/state.json","well_known_bundle":"https://pith.science/.well-known/pith/23IZB7K2H2STHS5OGU3SX45Y43/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:23IZB7K2H2STHS5OGU3SX45Y43","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":"b2d5a1ffd2f126b3687464bdd54878590cc00c30071a32d643bc8ba98db2121f","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-07-27T22:29:25Z","title_canon_sha256":"f64b8e348d873883aaed9189c75a52a859f0648b3f54a9627b55a100e94a8e16"},"schema_version":"1.0","source":{"id":"2307.15217","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.15217","created_at":"2026-05-17T23:39:21Z"},{"alias_kind":"arxiv_version","alias_value":"2307.15217v2","created_at":"2026-05-17T23:39:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.15217","created_at":"2026-05-17T23:39:21Z"},{"alias_kind":"pith_short_12","alias_value":"23IZB7K2H2ST","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"23IZB7K2H2STHS5O","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"23IZB7K2","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:4fde2cec5ebbdf1d6924a489b435f99e8ffc28db3a1b8cad8e2a461ba984061a","target":"graph","created_at":"2026-05-17T23:39:21Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"RLHF has emerged as the central method used to finetune state-of-the-art large language models but has fundamental limitations, emphasizing the importance of a multi-faceted approach to the development of safer AI systems."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the identified open problems represent fundamental limitations of RLHF rather than challenges that can be resolved through incremental improvements or better implementation."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"RLHF has significant open problems and fundamental limitations that require a multi-faceted approach for safer AI development."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"RLHF, the dominant method for aligning large language models with human goals, carries fundamental limitations that incremental fixes cannot fully resolve."}],"snapshot_sha256":"006f1672e3aad5d3409d2fe0f53106294ab14aa905c8916cb46d1259040d51e5"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"5803d0de03d454fe4fec6249f84c86bea3f67e278651388713b656953744c67d"},"paper":{"abstract_excerpt":"Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there has been relatively little public work systematizing its flaws. In this paper, we (1) survey open problems and fundamental limitations of RLHF and related methods; (2) overview techniques to understand, improve, and complement RLHF in practice; and (3) propose auditing and disclosure standards to improve societal oversight of RLHF systems. Our work","authors_text":"Anand Siththaranjan, Anca Dragan, Andi Peng, Charbel-Rapha\\\"el Segerie, Claudia Shi, David Krueger, David Lindner, Dmitrii Krasheninnikov, Dorsa Sadigh, Dylan Hadfield-Menell, Erdem B{\\i}y{\\i}k, Eric J. Michaud, Jacob Pfau, Javier Rando, J\\'er\\'emy Scheurer, Lauro Langosco, Max Nadeau, Mehul Damani, Micah Carroll, Pedro Freire, Peter Hase, Phillip Christoffersen, Rachel Freedman, Samuel Marks, Stephen Casper, Stewart Slocum, Thomas Krendl Gilbert, Tomasz Korbak, Tony Wang, Usman Anwar, Xander Davies, Xin Chen","cross_cats":["cs.CL","cs.LG"],"headline":"RLHF, the dominant method for aligning large language models with human goals, carries fundamental limitations that incremental fixes cannot fully resolve.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-07-27T22:29:25Z","title":"Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.15217","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-14T22:40:12.619170Z","id":"1928dcfc-c59b-4bc6-a627-8307e124f306","model_set":{"reader":"grok-4.3"},"one_line_summary":"RLHF has significant open problems and fundamental limitations that require a multi-faceted approach for safer AI development.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"RLHF, the dominant method for aligning large language models with human goals, carries fundamental limitations that incremental fixes cannot fully resolve.","strongest_claim":"RLHF has emerged as the central method used to finetune state-of-the-art large language models but has fundamental limitations, emphasizing the importance of a multi-faceted approach to the development of safer AI systems.","weakest_assumption":"That the identified open problems represent fundamental limitations of RLHF rather than challenges that can be resolved through incremental improvements or better implementation."}},"verdict_id":"1928dcfc-c59b-4bc6-a627-8307e124f306"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b62ab5e7af6944a12434dc205184fc460cdc7cfcadabd3ea5c37b08308edc198","target":"record","created_at":"2026-05-17T23:39:21Z","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":"b2d5a1ffd2f126b3687464bdd54878590cc00c30071a32d643bc8ba98db2121f","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-07-27T22:29:25Z","title_canon_sha256":"f64b8e348d873883aaed9189c75a52a859f0648b3f54a9627b55a100e94a8e16"},"schema_version":"1.0","source":{"id":"2307.15217","kind":"arxiv","version":2}},"canonical_sha256":"d6d190fd5a3ea533cbae35372bf3b8e6ddf630437eb272f01b14c5a437f010e0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d6d190fd5a3ea533cbae35372bf3b8e6ddf630437eb272f01b14c5a437f010e0","first_computed_at":"2026-05-17T23:39:21.418541Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:21.418541Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IgWYAzu8E11YJKSnEBOZRO42bFde2+JBCzbLVCW8GfzIcKwkkklGwB+qS0rz/wj0+t86Er3LpC7yetHSWlyQCQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:21.419160Z","signed_message":"canonical_sha256_bytes"},"source_id":"2307.15217","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b62ab5e7af6944a12434dc205184fc460cdc7cfcadabd3ea5c37b08308edc198","sha256:4fde2cec5ebbdf1d6924a489b435f99e8ffc28db3a1b8cad8e2a461ba984061a"],"state_sha256":"08d9cae4d0f78607ffd8dfb0d876a83e3dadb724096f76136014a8b3a5fd10fe"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3thH4HrULjQ0wELNAaCteZwgkN5nQfsz2IrdOkqcTz4W2YSrACAH3ATYbEGycz+a9UvSBefXliP+nL37J1GaBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T16:56:19.596634Z","bundle_sha256":"871528c3953d0cb7d75ff4a157068ebfcaad9a69b88ed4c982dd0682a54e018c"}}