{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GLIPD37OQ23SYYLFHTLMTCM6LS","short_pith_number":"pith:GLIPD37O","canonical_record":{"source":{"id":"2404.03715","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-04T17:56:41Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"447afa09a2785d3f245dab4ca48accd02be554af48c1dcc6a4630ffdd7e4f5f4","abstract_canon_sha256":"8b2a61351b37757790193007ca2c4d424a4718c88bef6a32ad4d55ef26930cb8"},"schema_version":"1.0"},"canonical_sha256":"32d0f1efee86b72c61653cd6c9899e5c88c34bb50098d6f7a1f38aaf926c6ad4","source":{"kind":"arxiv","id":"2404.03715","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.03715","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"arxiv_version","alias_value":"2404.03715v1","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03715","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"pith_short_12","alias_value":"GLIPD37OQ23S","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"pith_short_16","alias_value":"GLIPD37OQ23SYYLF","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"pith_short_8","alias_value":"GLIPD37O","created_at":"2026-07-05T08:04:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GLIPD37OQ23SYYLFHTLMTCM6LS","target":"record","payload":{"canonical_record":{"source":{"id":"2404.03715","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-04T17:56:41Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"447afa09a2785d3f245dab4ca48accd02be554af48c1dcc6a4630ffdd7e4f5f4","abstract_canon_sha256":"8b2a61351b37757790193007ca2c4d424a4718c88bef6a32ad4d55ef26930cb8"},"schema_version":"1.0"},"canonical_sha256":"32d0f1efee86b72c61653cd6c9899e5c88c34bb50098d6f7a1f38aaf926c6ad4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:04:40.584475Z","signature_b64":"DYBBa7WM4NtVa6DiBhDzrx0GRQQEKgbAxb4co0X58dxYETMDac1Yp5fU3q7Ro+WI1wvR/EdqFqHWV9Ohbm40Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32d0f1efee86b72c61653cd6c9899e5c88c34bb50098d6f7a1f38aaf926c6ad4","last_reissued_at":"2026-07-05T08:04:40.583993Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:04:40.583993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.03715","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":"2026-07-05T08:04:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+HVorXx/N7OALVBcDX0HuoJq9wYKJo/wEhKlmOfF1CVjAPhoX+bXyhKA6k2KJltN74o5y5Yt8Ok6GA0JgZVAAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T05:53:38.720944Z"},"content_sha256":"03222a2b0ec98d7e03ac292df8461e2594bb4ec0ff3612742d69fc6bfdd6c7d0","schema_version":"1.0","event_id":"sha256:03222a2b0ec98d7e03ac292df8461e2594bb4ec0ff3612742d69fc6bfdd6c7d0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GLIPD37OQ23SYYLFHTLMTCM6LS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Ahmed Awadallah, Arindam Mitra, Ching-An Cheng, Corby Rosset, Michael Santacroce, Tengyang Xie","submitted_at":"2024-04-04T17:56:41Z","abstract_excerpt":"This paper studies post-training large language models (LLMs) using preference feedback from a powerful oracle to help a model iteratively improve over itself. The typical approach for post-training LLMs involves Reinforcement Learning from Human Feedback (RLHF), which traditionally separates reward learning and subsequent policy optimization. However, such a reward maximization approach is limited by the nature of \"point-wise\" rewards (such as Bradley-Terry model), which fails to express complex intransitive or cyclic preference relations. While advances on RLHF show reward learning and polic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03715","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/2404.03715/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":"2026-07-05T08:04:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kf/BkvaLuScvlZxTkrK/CC4FvnMOjCWYbUJbBgJbEGff0+opHjvvqlwdMBb7mkfId6IgtraNbc2Yk39N9s3cCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T05:53:38.721610Z"},"content_sha256":"e456cbe1495d23f992400db31d43e7e9d518131bf92040b4bfa68b8762353f75","schema_version":"1.0","event_id":"sha256:e456cbe1495d23f992400db31d43e7e9d518131bf92040b4bfa68b8762353f75"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GLIPD37OQ23SYYLFHTLMTCM6LS/bundle.json","state_url":"https://pith.science/pith/GLIPD37OQ23SYYLFHTLMTCM6LS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GLIPD37OQ23SYYLFHTLMTCM6LS/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":"2026-08-04T05:53:38Z","links":{"resolver":"https://pith.science/pith/GLIPD37OQ23SYYLFHTLMTCM6LS","bundle":"https://pith.science/pith/GLIPD37OQ23SYYLFHTLMTCM6LS/bundle.json","state":"https://pith.science/pith/GLIPD37OQ23SYYLFHTLMTCM6LS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GLIPD37OQ23SYYLFHTLMTCM6LS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GLIPD37OQ23SYYLFHTLMTCM6LS","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":"8b2a61351b37757790193007ca2c4d424a4718c88bef6a32ad4d55ef26930cb8","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-04T17:56:41Z","title_canon_sha256":"447afa09a2785d3f245dab4ca48accd02be554af48c1dcc6a4630ffdd7e4f5f4"},"schema_version":"1.0","source":{"id":"2404.03715","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.03715","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"arxiv_version","alias_value":"2404.03715v1","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03715","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"pith_short_12","alias_value":"GLIPD37OQ23S","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"pith_short_16","alias_value":"GLIPD37OQ23SYYLF","created_at":"2026-07-05T08:04:40Z"},{"alias_kind":"pith_short_8","alias_value":"GLIPD37O","created_at":"2026-07-05T08:04:40Z"}],"graph_snapshots":[{"event_id":"sha256:e456cbe1495d23f992400db31d43e7e9d518131bf92040b4bfa68b8762353f75","target":"graph","created_at":"2026-07-05T08:04:40Z","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/2404.03715/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper studies post-training large language models (LLMs) using preference feedback from a powerful oracle to help a model iteratively improve over itself. The typical approach for post-training LLMs involves Reinforcement Learning from Human Feedback (RLHF), which traditionally separates reward learning and subsequent policy optimization. However, such a reward maximization approach is limited by the nature of \"point-wise\" rewards (such as Bradley-Terry model), which fails to express complex intransitive or cyclic preference relations. While advances on RLHF show reward learning and polic","authors_text":"Ahmed Awadallah, Arindam Mitra, Ching-An Cheng, Corby Rosset, Michael Santacroce, Tengyang Xie","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-04T17:56:41Z","title":"Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03715","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":"sha256:03222a2b0ec98d7e03ac292df8461e2594bb4ec0ff3612742d69fc6bfdd6c7d0","target":"record","created_at":"2026-07-05T08:04:40Z","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":"8b2a61351b37757790193007ca2c4d424a4718c88bef6a32ad4d55ef26930cb8","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-04T17:56:41Z","title_canon_sha256":"447afa09a2785d3f245dab4ca48accd02be554af48c1dcc6a4630ffdd7e4f5f4"},"schema_version":"1.0","source":{"id":"2404.03715","kind":"arxiv","version":1}},"canonical_sha256":"32d0f1efee86b72c61653cd6c9899e5c88c34bb50098d6f7a1f38aaf926c6ad4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"32d0f1efee86b72c61653cd6c9899e5c88c34bb50098d6f7a1f38aaf926c6ad4","first_computed_at":"2026-07-05T08:04:40.583993Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:04:40.583993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DYBBa7WM4NtVa6DiBhDzrx0GRQQEKgbAxb4co0X58dxYETMDac1Yp5fU3q7Ro+WI1wvR/EdqFqHWV9Ohbm40Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:04:40.584475Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.03715","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:03222a2b0ec98d7e03ac292df8461e2594bb4ec0ff3612742d69fc6bfdd6c7d0","sha256:e456cbe1495d23f992400db31d43e7e9d518131bf92040b4bfa68b8762353f75"],"state_sha256":"15b6710d8bfe5b490046f8722dbe01c517828da0026f189c8c258183490f78fa"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w2hVtGmJ5DuRXDexdmuoM6Q/IIYxuGakUfb0ip6NZ0rdMCI2Zn2cIBwo51JvGxaYop2It2B6baZbXwyhZkQwBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T05:53:38.728410Z","bundle_sha256":"e26a94aa097dd9bd626f9a40e1de280d3cb7dec5327587ca497d179ad18080ff"}}