{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:F5IT2WWDUMTDIXHS6B526KQ4PM","short_pith_number":"pith:F5IT2WWD","canonical_record":{"source":{"id":"2402.14979","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-22T21:36:07Z","cross_cats_sorted":["cs.CL","stat.ME"],"title_canon_sha256":"2dadf2265b49da20936f84453684d1072174d61d42368519840f95e3ecd7b27d","abstract_canon_sha256":"24672aa3ed663a0ea079b3761bb154cd7729dbb2a62fec715efcd57f5ce0154b"},"schema_version":"1.0"},"canonical_sha256":"2f513d5ac3a326345cf2f07baf2a1c7b25f0779788ca8e578771caffcf629f1b","source":{"kind":"arxiv","id":"2402.14979","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.14979","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"arxiv_version","alias_value":"2402.14979v2","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14979","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"pith_short_12","alias_value":"F5IT2WWDUMTD","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"pith_short_16","alias_value":"F5IT2WWDUMTDIXHS","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"pith_short_8","alias_value":"F5IT2WWD","created_at":"2026-07-05T08:28:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:F5IT2WWDUMTDIXHS6B526KQ4PM","target":"record","payload":{"canonical_record":{"source":{"id":"2402.14979","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-22T21:36:07Z","cross_cats_sorted":["cs.CL","stat.ME"],"title_canon_sha256":"2dadf2265b49da20936f84453684d1072174d61d42368519840f95e3ecd7b27d","abstract_canon_sha256":"24672aa3ed663a0ea079b3761bb154cd7729dbb2a62fec715efcd57f5ce0154b"},"schema_version":"1.0"},"canonical_sha256":"2f513d5ac3a326345cf2f07baf2a1c7b25f0779788ca8e578771caffcf629f1b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:06.809507Z","signature_b64":"9mdJ2GFFwISvvCDwBb34y81QSXvy3AvhuJrBd797XoIjt9+9TvqdDR97ion+rL2jvYi4hW/UNZYb9UlHzSbmDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f513d5ac3a326345cf2f07baf2a1c7b25f0779788ca8e578771caffcf629f1b","last_reissued_at":"2026-07-05T08:28:06.809035Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:06.809035Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.14979","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":"2026-07-05T08:28:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Vn1IsW4JAXpPFC3TgSGgK3y87xfmllTDaKrLHHkaOwXfiPcGc32EVb9XymEUWx3JiIibGGzebLFA8RDbsFd6Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:26:54.091357Z"},"content_sha256":"d9401f49034aec43cef008c4312e30d331836a53dd0555f1c95b1d23db9bb173","schema_version":"1.0","event_id":"sha256:d9401f49034aec43cef008c4312e30d331836a53dd0555f1c95b1d23db9bb173"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:F5IT2WWDUMTDIXHS6B526KQ4PM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimizing Language Models for Human Preferences is a Causal Inference Problem","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","stat.ME"],"primary_cat":"cs.LG","authors_text":"Eli Ben-Michael, Louis-Philippe Morency, Victoria Lin","submitted_at":"2024-02-22T21:36:07Z","abstract_excerpt":"As large language models (LLMs) see greater use in academic and commercial settings, there is increasing interest in methods that allow language models to generate texts aligned with human preferences. In this paper, we present an initial exploration of language model optimization for human preferences from direct outcome datasets, where each sample consists of a text and an associated numerical outcome measuring the reader's response. We first propose that language model optimization should be viewed as a causal problem to ensure that the model correctly learns the relationship between the te"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14979","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/2402.14979/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":"2026-07-05T08:28:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LpL9vL8xaF9SMDCMvSK2DRgf40CJ9PuXnS5zNLa64d0JI/9WeYeCiutdKwe9WqQw7BoUHl2wnEFiypiuktk6Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:26:54.091738Z"},"content_sha256":"6563db2ea31fb6e27adc609d7613e9a75937e66c1050c6bbdcdf149a8c200831","schema_version":"1.0","event_id":"sha256:6563db2ea31fb6e27adc609d7613e9a75937e66c1050c6bbdcdf149a8c200831"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/F5IT2WWDUMTDIXHS6B526KQ4PM/bundle.json","state_url":"https://pith.science/pith/F5IT2WWDUMTDIXHS6B526KQ4PM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/F5IT2WWDUMTDIXHS6B526KQ4PM/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":"2026-08-07T18:26:54Z","links":{"resolver":"https://pith.science/pith/F5IT2WWDUMTDIXHS6B526KQ4PM","bundle":"https://pith.science/pith/F5IT2WWDUMTDIXHS6B526KQ4PM/bundle.json","state":"https://pith.science/pith/F5IT2WWDUMTDIXHS6B526KQ4PM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/F5IT2WWDUMTDIXHS6B526KQ4PM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:F5IT2WWDUMTDIXHS6B526KQ4PM","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":"24672aa3ed663a0ea079b3761bb154cd7729dbb2a62fec715efcd57f5ce0154b","cross_cats_sorted":["cs.CL","stat.ME"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-22T21:36:07Z","title_canon_sha256":"2dadf2265b49da20936f84453684d1072174d61d42368519840f95e3ecd7b27d"},"schema_version":"1.0","source":{"id":"2402.14979","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.14979","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"arxiv_version","alias_value":"2402.14979v2","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14979","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"pith_short_12","alias_value":"F5IT2WWDUMTD","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"pith_short_16","alias_value":"F5IT2WWDUMTDIXHS","created_at":"2026-07-05T08:28:06Z"},{"alias_kind":"pith_short_8","alias_value":"F5IT2WWD","created_at":"2026-07-05T08:28:06Z"}],"graph_snapshots":[{"event_id":"sha256:6563db2ea31fb6e27adc609d7613e9a75937e66c1050c6bbdcdf149a8c200831","target":"graph","created_at":"2026-07-05T08:28:06Z","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/2402.14979/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) see greater use in academic and commercial settings, there is increasing interest in methods that allow language models to generate texts aligned with human preferences. In this paper, we present an initial exploration of language model optimization for human preferences from direct outcome datasets, where each sample consists of a text and an associated numerical outcome measuring the reader's response. We first propose that language model optimization should be viewed as a causal problem to ensure that the model correctly learns the relationship between the te","authors_text":"Eli Ben-Michael, Louis-Philippe Morency, Victoria Lin","cross_cats":["cs.CL","stat.ME"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-22T21:36:07Z","title":"Optimizing Language Models for Human Preferences is a Causal Inference Problem"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14979","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:d9401f49034aec43cef008c4312e30d331836a53dd0555f1c95b1d23db9bb173","target":"record","created_at":"2026-07-05T08:28:06Z","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":"24672aa3ed663a0ea079b3761bb154cd7729dbb2a62fec715efcd57f5ce0154b","cross_cats_sorted":["cs.CL","stat.ME"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-22T21:36:07Z","title_canon_sha256":"2dadf2265b49da20936f84453684d1072174d61d42368519840f95e3ecd7b27d"},"schema_version":"1.0","source":{"id":"2402.14979","kind":"arxiv","version":2}},"canonical_sha256":"2f513d5ac3a326345cf2f07baf2a1c7b25f0779788ca8e578771caffcf629f1b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2f513d5ac3a326345cf2f07baf2a1c7b25f0779788ca8e578771caffcf629f1b","first_computed_at":"2026-07-05T08:28:06.809035Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:28:06.809035Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9mdJ2GFFwISvvCDwBb34y81QSXvy3AvhuJrBd797XoIjt9+9TvqdDR97ion+rL2jvYi4hW/UNZYb9UlHzSbmDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:28:06.809507Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.14979","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d9401f49034aec43cef008c4312e30d331836a53dd0555f1c95b1d23db9bb173","sha256:6563db2ea31fb6e27adc609d7613e9a75937e66c1050c6bbdcdf149a8c200831"],"state_sha256":"87eef8279790fc1f7f66198df45df12f498183ecd9b87b21ab30ecf86bd36b77"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ndbQWbh4/bIlBktTu56HNcIdX6WhO9Kn8w00Yx5971fyQpN1Wr/JE4p+09H00qhP1hQucHjkQMMHapIX7c8cCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T18:26:54.094976Z","bundle_sha256":"974af75e7b49d6aeb6e5614422b5428cae6b5ced8115f55e83393ec6459d39b1"}}