{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:BAMYOO6Y2IFX2SUMWISKAG3ZKP","short_pith_number":"pith:BAMYOO6Y","canonical_record":{"source":{"id":"2310.16763","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-25T16:52:00Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"1a4592d41ca438c5a71b1f0539e2a51a23d170eb9b0101f28f5d3d02d37cd9b2","abstract_canon_sha256":"0f8d46cd03dad86b23af08fd66171ad0aec1b33449022839bb832a295aca2d2d"},"schema_version":"1.0"},"canonical_sha256":"0819873bd8d20b7d4a8cb224a01b7953f76208c3e9210551408dea82938e6eaf","source":{"kind":"arxiv","id":"2310.16763","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.16763","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"arxiv_version","alias_value":"2310.16763v1","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.16763","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_12","alias_value":"BAMYOO6Y2IFX","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_16","alias_value":"BAMYOO6Y2IFX2SUM","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_8","alias_value":"BAMYOO6Y","created_at":"2026-07-05T07:05:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:BAMYOO6Y2IFX2SUMWISKAG3ZKP","target":"record","payload":{"canonical_record":{"source":{"id":"2310.16763","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-25T16:52:00Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"1a4592d41ca438c5a71b1f0539e2a51a23d170eb9b0101f28f5d3d02d37cd9b2","abstract_canon_sha256":"0f8d46cd03dad86b23af08fd66171ad0aec1b33449022839bb832a295aca2d2d"},"schema_version":"1.0"},"canonical_sha256":"0819873bd8d20b7d4a8cb224a01b7953f76208c3e9210551408dea82938e6eaf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:05:02.750737Z","signature_b64":"zm+YZJ8047/8KYG1KPZbqfJ0wzftwSiIF4QyH10Q4hzUp9iYkMr6NdQrjdO/X0tnAHg5h0cD4VCM5UPnNlSICQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0819873bd8d20b7d4a8cb224a01b7953f76208c3e9210551408dea82938e6eaf","last_reissued_at":"2026-07-05T07:05:02.750209Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:05:02.750209Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.16763","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:05:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZO5fr5JVXeqSrWqtvYAHkXCUm2/c1kM91VzfPfGgv2cKzRjp2t9kHuOU5OutP0E9Pre25YpYX91bvy1HY/LLAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T22:30:47.483718Z"},"content_sha256":"51641371c72cf2bc75ea8c5ed8f44026c1ca15237ef55f61524579d7d93b327d","schema_version":"1.0","event_id":"sha256:51641371c72cf2bc75ea8c5ed8f44026c1ca15237ef55f61524579d7d93b327d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:BAMYOO6Y2IFX2SUMWISKAG3ZKP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SuperHF: Supervised Iterative Learning from Human Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Gabriel Mukobi, Gitta Kutyniok, Kush Bhatia, Peter Chatain, Robert Windesheim, Silas Alberti, Su Fong","submitted_at":"2023-10-25T16:52:00Z","abstract_excerpt":"While large language models demonstrate remarkable capabilities, they often present challenges in terms of safety, alignment with human values, and stability during training. Here, we focus on two prevalent methods used to align these models, Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF). SFT is simple and robust, powering a host of open-source models, while RLHF is a more sophisticated method used in top-tier models like ChatGPT but also suffers from instability and susceptibility to reward hacking. We propose a novel approach, Supervised Iterative Learnin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.16763","kind":"arxiv","version":1},"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/2310.16763/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":"2026-07-05T07:05:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yETMvlLvhxZq3PCEmlk5L1yDaqXTH7kkaVBxFrYPfMOMlIVRuVzjpWLL1itnm85HokdFWTNA6DFlPeirdAEjDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T22:30:47.484343Z"},"content_sha256":"c12b4542105be513e5dad6ef1901bd05b68cb57f36ddba2f4fc9729cbcb10805","schema_version":"1.0","event_id":"sha256:c12b4542105be513e5dad6ef1901bd05b68cb57f36ddba2f4fc9729cbcb10805"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BAMYOO6Y2IFX2SUMWISKAG3ZKP/bundle.json","state_url":"https://pith.science/pith/BAMYOO6Y2IFX2SUMWISKAG3ZKP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BAMYOO6Y2IFX2SUMWISKAG3ZKP/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":"2026-07-31T22:30:47Z","links":{"resolver":"https://pith.science/pith/BAMYOO6Y2IFX2SUMWISKAG3ZKP","bundle":"https://pith.science/pith/BAMYOO6Y2IFX2SUMWISKAG3ZKP/bundle.json","state":"https://pith.science/pith/BAMYOO6Y2IFX2SUMWISKAG3ZKP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BAMYOO6Y2IFX2SUMWISKAG3ZKP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:BAMYOO6Y2IFX2SUMWISKAG3ZKP","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":"0f8d46cd03dad86b23af08fd66171ad0aec1b33449022839bb832a295aca2d2d","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-25T16:52:00Z","title_canon_sha256":"1a4592d41ca438c5a71b1f0539e2a51a23d170eb9b0101f28f5d3d02d37cd9b2"},"schema_version":"1.0","source":{"id":"2310.16763","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.16763","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"arxiv_version","alias_value":"2310.16763v1","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.16763","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_12","alias_value":"BAMYOO6Y2IFX","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_16","alias_value":"BAMYOO6Y2IFX2SUM","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_8","alias_value":"BAMYOO6Y","created_at":"2026-07-05T07:05:02Z"}],"graph_snapshots":[{"event_id":"sha256:c12b4542105be513e5dad6ef1901bd05b68cb57f36ddba2f4fc9729cbcb10805","target":"graph","created_at":"2026-07-05T07:05:02Z","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/2310.16763/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While large language models demonstrate remarkable capabilities, they often present challenges in terms of safety, alignment with human values, and stability during training. Here, we focus on two prevalent methods used to align these models, Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF). SFT is simple and robust, powering a host of open-source models, while RLHF is a more sophisticated method used in top-tier models like ChatGPT but also suffers from instability and susceptibility to reward hacking. We propose a novel approach, Supervised Iterative Learnin","authors_text":"Gabriel Mukobi, Gitta Kutyniok, Kush Bhatia, Peter Chatain, Robert Windesheim, Silas Alberti, Su Fong","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-25T16:52:00Z","title":"SuperHF: Supervised Iterative Learning from Human Feedback"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.16763","kind":"arxiv","version":1},"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":"sha256:51641371c72cf2bc75ea8c5ed8f44026c1ca15237ef55f61524579d7d93b327d","target":"record","created_at":"2026-07-05T07:05:02Z","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":"0f8d46cd03dad86b23af08fd66171ad0aec1b33449022839bb832a295aca2d2d","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-25T16:52:00Z","title_canon_sha256":"1a4592d41ca438c5a71b1f0539e2a51a23d170eb9b0101f28f5d3d02d37cd9b2"},"schema_version":"1.0","source":{"id":"2310.16763","kind":"arxiv","version":1}},"canonical_sha256":"0819873bd8d20b7d4a8cb224a01b7953f76208c3e9210551408dea82938e6eaf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0819873bd8d20b7d4a8cb224a01b7953f76208c3e9210551408dea82938e6eaf","first_computed_at":"2026-07-05T07:05:02.750209Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:05:02.750209Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zm+YZJ8047/8KYG1KPZbqfJ0wzftwSiIF4QyH10Q4hzUp9iYkMr6NdQrjdO/X0tnAHg5h0cD4VCM5UPnNlSICQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:05:02.750737Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.16763","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:51641371c72cf2bc75ea8c5ed8f44026c1ca15237ef55f61524579d7d93b327d","sha256:c12b4542105be513e5dad6ef1901bd05b68cb57f36ddba2f4fc9729cbcb10805"],"state_sha256":"91d81bfa3a306bb5a6851a240375e34ebe82de83c8e99a2138d3f980e9bf3d9a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LKJ1kwmUixYhxJUbLZimq+R7Jmbp5guUNufq4pipTkD/D17oe5nZ+IZCje76O8Mm7lYrkBb71qv8y1/i9JgKCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T22:30:47.489524Z","bundle_sha256":"b7652ae33edca78a9250755ec7469bb604888ed34dd64707f7e76751e58b3582"}}