{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:33H6ZU7C6OE3QQOC6GTPMYVL7Z","short_pith_number":"pith:33H6ZU7C","canonical_record":{"source":{"id":"2401.16635","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-30T00:17:37Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"baf25723ac96a73175e1437debd69bbd55ff753535253dd0d4023b1d9f09061a","abstract_canon_sha256":"5ac83861de289a1a0924ac6ce5e749593aa7693640f554f28e7d842de265c882"},"schema_version":"1.0"},"canonical_sha256":"decfecd3e2f389b841c2f1a6f662abfe507d9113b719c46c6246ee609b681cec","source":{"kind":"arxiv","id":"2401.16635","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.16635","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"arxiv_version","alias_value":"2401.16635v3","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.16635","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"pith_short_12","alias_value":"33H6ZU7C6OE3","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"pith_short_16","alias_value":"33H6ZU7C6OE3QQOC","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"pith_short_8","alias_value":"33H6ZU7C","created_at":"2026-07-05T09:23:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:33H6ZU7C6OE3QQOC6GTPMYVL7Z","target":"record","payload":{"canonical_record":{"source":{"id":"2401.16635","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-30T00:17:37Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"baf25723ac96a73175e1437debd69bbd55ff753535253dd0d4023b1d9f09061a","abstract_canon_sha256":"5ac83861de289a1a0924ac6ce5e749593aa7693640f554f28e7d842de265c882"},"schema_version":"1.0"},"canonical_sha256":"decfecd3e2f389b841c2f1a6f662abfe507d9113b719c46c6246ee609b681cec","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:23:46.684165Z","signature_b64":"ej6W07qtuBNQdj+acA80vnnt5hHZr7ZlRNjXU/2cqFa/ooxxbFpB9q4S9C3MpUbM3Hm2An/uxJfytWKrvBCVCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"decfecd3e2f389b841c2f1a6f662abfe507d9113b719c46c6246ee609b681cec","last_reissued_at":"2026-07-05T09:23:46.683638Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:23:46.683638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.16635","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:23:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RfaPmGU+gkyj4qTHZuMfs7iRV/ywiyyIs8EHpNDgehOujwK2qUum0QgkDjr2EzEyvkyI8YGgO/PuznaxBepKDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:42:31.814921Z"},"content_sha256":"9028de35a840c3d063b35eb0efdecc1b26126d979b29725e0cbabdd00bbc2f8b","schema_version":"1.0","event_id":"sha256:9028de35a840c3d063b35eb0efdecc1b26126d979b29725e0cbabdd00bbc2f8b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:33H6ZU7C6OE3QQOC6GTPMYVL7Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving Reinforcement Learning from Human Feedback with Efficient Reward Model Ensemble","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Chuang Gan, Shun Zhang, Sunli Chen, Yikang Shen, Zhenfang Chen, Zhiqing Sun","submitted_at":"2024-01-30T00:17:37Z","abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) is a widely adopted approach for aligning large language models with human values. However, RLHF relies on a reward model that is trained with a limited amount of human preference data, which could lead to inaccurate predictions. As a result, RLHF may produce outputs that are misaligned with human values. To mitigate this issue, we contribute a reward ensemble method that allows the reward model to make more accurate predictions. As using an ensemble of large language model-based reward models can be computationally and resource-expensive, we e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.16635","kind":"arxiv","version":3},"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/2401.16635/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":"2026-07-05T09:23:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rwdwwI685rrDqmHVKXIE/gqiEez9DMoW00B5wVyucHvLKY6fP4mI4cZX8wxqRJDamu8RUEtp6z3890ri33UhDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:42:31.815303Z"},"content_sha256":"dec033f2c24d27de8f89fb0c80db4d049b5a4cd63f286d6d52906779f6e1066e","schema_version":"1.0","event_id":"sha256:dec033f2c24d27de8f89fb0c80db4d049b5a4cd63f286d6d52906779f6e1066e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/33H6ZU7C6OE3QQOC6GTPMYVL7Z/bundle.json","state_url":"https://pith.science/pith/33H6ZU7C6OE3QQOC6GTPMYVL7Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/33H6ZU7C6OE3QQOC6GTPMYVL7Z/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":"2026-07-07T09:42:31Z","links":{"resolver":"https://pith.science/pith/33H6ZU7C6OE3QQOC6GTPMYVL7Z","bundle":"https://pith.science/pith/33H6ZU7C6OE3QQOC6GTPMYVL7Z/bundle.json","state":"https://pith.science/pith/33H6ZU7C6OE3QQOC6GTPMYVL7Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/33H6ZU7C6OE3QQOC6GTPMYVL7Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:33H6ZU7C6OE3QQOC6GTPMYVL7Z","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":"5ac83861de289a1a0924ac6ce5e749593aa7693640f554f28e7d842de265c882","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-30T00:17:37Z","title_canon_sha256":"baf25723ac96a73175e1437debd69bbd55ff753535253dd0d4023b1d9f09061a"},"schema_version":"1.0","source":{"id":"2401.16635","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.16635","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"arxiv_version","alias_value":"2401.16635v3","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.16635","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"pith_short_12","alias_value":"33H6ZU7C6OE3","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"pith_short_16","alias_value":"33H6ZU7C6OE3QQOC","created_at":"2026-07-05T09:23:46Z"},{"alias_kind":"pith_short_8","alias_value":"33H6ZU7C","created_at":"2026-07-05T09:23:46Z"}],"graph_snapshots":[{"event_id":"sha256:dec033f2c24d27de8f89fb0c80db4d049b5a4cd63f286d6d52906779f6e1066e","target":"graph","created_at":"2026-07-05T09:23:46Z","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/2401.16635/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) is a widely adopted approach for aligning large language models with human values. However, RLHF relies on a reward model that is trained with a limited amount of human preference data, which could lead to inaccurate predictions. As a result, RLHF may produce outputs that are misaligned with human values. To mitigate this issue, we contribute a reward ensemble method that allows the reward model to make more accurate predictions. As using an ensemble of large language model-based reward models can be computationally and resource-expensive, we e","authors_text":"Chuang Gan, Shun Zhang, Sunli Chen, Yikang Shen, Zhenfang Chen, Zhiqing Sun","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-30T00:17:37Z","title":"Improving Reinforcement Learning from Human Feedback with Efficient Reward Model Ensemble"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.16635","kind":"arxiv","version":3},"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":"sha256:9028de35a840c3d063b35eb0efdecc1b26126d979b29725e0cbabdd00bbc2f8b","target":"record","created_at":"2026-07-05T09:23:46Z","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":"5ac83861de289a1a0924ac6ce5e749593aa7693640f554f28e7d842de265c882","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-30T00:17:37Z","title_canon_sha256":"baf25723ac96a73175e1437debd69bbd55ff753535253dd0d4023b1d9f09061a"},"schema_version":"1.0","source":{"id":"2401.16635","kind":"arxiv","version":3}},"canonical_sha256":"decfecd3e2f389b841c2f1a6f662abfe507d9113b719c46c6246ee609b681cec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"decfecd3e2f389b841c2f1a6f662abfe507d9113b719c46c6246ee609b681cec","first_computed_at":"2026-07-05T09:23:46.683638Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:23:46.683638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ej6W07qtuBNQdj+acA80vnnt5hHZr7ZlRNjXU/2cqFa/ooxxbFpB9q4S9C3MpUbM3Hm2An/uxJfytWKrvBCVCg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:23:46.684165Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.16635","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9028de35a840c3d063b35eb0efdecc1b26126d979b29725e0cbabdd00bbc2f8b","sha256:dec033f2c24d27de8f89fb0c80db4d049b5a4cd63f286d6d52906779f6e1066e"],"state_sha256":"0aac8efbd867a649a7f0f3dc61581b925c142e226398b217f8ac7d41dbc59002"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XzY+tbVsJ/+trYiokdBxXLrkLWTbb5VV9hEoEHWtq0TbHLTm787CZjqkI4PfuiwoUEIK6DjDBjiP9GgUE6oyCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T09:42:31.817661Z","bundle_sha256":"cd48a2178b50e3f807fbd9b03d170376793ced4327a1d11227b7643cdfb73465"}}