{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:NDMPVL2LZO4LCFDXTFCXSL7OWX","short_pith_number":"pith:NDMPVL2L","canonical_record":{"source":{"id":"2411.11937","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-18T16:12:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7c566c6c5dc2d32c21eb921540ccc70bf2fe9af35d1b067377399618953eeee9","abstract_canon_sha256":"da4500aeb5d01374b708685c0e8b79c6a1fccd65d3b30682caa66a4b32e1c58a"},"schema_version":"1.0"},"canonical_sha256":"68d8faaf4bcbb8b114779945792feeb5c0dc038b3ad7777674d7ae4a4be53f9f","source":{"kind":"arxiv","id":"2411.11937","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.11937","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"arxiv_version","alias_value":"2411.11937v1","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.11937","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"pith_short_12","alias_value":"NDMPVL2LZO4L","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"pith_short_16","alias_value":"NDMPVL2LZO4LCFDX","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"pith_short_8","alias_value":"NDMPVL2L","created_at":"2026-07-05T09:37:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:NDMPVL2LZO4LCFDXTFCXSL7OWX","target":"record","payload":{"canonical_record":{"source":{"id":"2411.11937","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-18T16:12:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7c566c6c5dc2d32c21eb921540ccc70bf2fe9af35d1b067377399618953eeee9","abstract_canon_sha256":"da4500aeb5d01374b708685c0e8b79c6a1fccd65d3b30682caa66a4b32e1c58a"},"schema_version":"1.0"},"canonical_sha256":"68d8faaf4bcbb8b114779945792feeb5c0dc038b3ad7777674d7ae4a4be53f9f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:37:14.139961Z","signature_b64":"GwlEkI/H0gy/LvWro7eeIE1mE7bK36cRolGgA7R+K2w7ABMOhDv7wRKI3D0462fTIWw71Awtv9diLi4gQcXOCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"68d8faaf4bcbb8b114779945792feeb5c0dc038b3ad7777674d7ae4a4be53f9f","last_reissued_at":"2026-07-05T09:37:14.139341Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:37:14.139341Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.11937","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":"2026-07-05T09:37:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bzMieMEL94+bcaXp73KkCgW7D3o2IaXMzopPcgQIUb8M5ykSwy6V67dI5tHD5LafVS/xDP7aCwNKfablKuX6Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T16:57:46.495221Z"},"content_sha256":"a0a2d839c9f994048e341974984c59ce368e7d11f3b0431dc0d2247ad7b33ac7","schema_version":"1.0","event_id":"sha256:a0a2d839c9f994048e341974984c59ce368e7d11f3b0431dc0d2247ad7b33ac7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:NDMPVL2LZO4LCFDXTFCXSL7OWX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Value Imprint: A Technique for Auditing the Human Values Embedded in RLHF Datasets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aaron Basiletti, Ike Obi, Maham Ghazanfar, Rohan Pant, Srishti Shekhar Agrawal","submitted_at":"2024-11-18T16:12:24Z","abstract_excerpt":"LLMs are increasingly fine-tuned using RLHF datasets to align them with human preferences and values. However, very limited research has investigated which specific human values are operationalized through these datasets. In this paper, we introduce Value Imprint, a framework for auditing and classifying the human values embedded within RLHF datasets. To investigate the viability of this framework, we conducted three case study experiments by auditing the Anthropic/hh-rlhf, OpenAI WebGPT Comparisons, and Alpaca GPT-4-LLM datasets to examine the human values embedded within them. Our analysis i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.11937","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/2411.11937/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":"2026-07-05T09:37:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"58ClM5abpoafpvN0YkDG8ouwyhc8JT8BbHaQkvHXNmfdp3OKHyMoNxZWqQy0Jk72oAvKlIiAJ1iIhe+27zDRAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T16:57:46.496481Z"},"content_sha256":"a0a80e4330debbb48606b8503b48b9b1071368141cf0f84d348b20038e3915b1","schema_version":"1.0","event_id":"sha256:a0a80e4330debbb48606b8503b48b9b1071368141cf0f84d348b20038e3915b1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NDMPVL2LZO4LCFDXTFCXSL7OWX/bundle.json","state_url":"https://pith.science/pith/NDMPVL2LZO4LCFDXTFCXSL7OWX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NDMPVL2LZO4LCFDXTFCXSL7OWX/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":"2026-08-13T16:57:46Z","links":{"resolver":"https://pith.science/pith/NDMPVL2LZO4LCFDXTFCXSL7OWX","bundle":"https://pith.science/pith/NDMPVL2LZO4LCFDXTFCXSL7OWX/bundle.json","state":"https://pith.science/pith/NDMPVL2LZO4LCFDXTFCXSL7OWX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NDMPVL2LZO4LCFDXTFCXSL7OWX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NDMPVL2LZO4LCFDXTFCXSL7OWX","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":"da4500aeb5d01374b708685c0e8b79c6a1fccd65d3b30682caa66a4b32e1c58a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-18T16:12:24Z","title_canon_sha256":"7c566c6c5dc2d32c21eb921540ccc70bf2fe9af35d1b067377399618953eeee9"},"schema_version":"1.0","source":{"id":"2411.11937","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.11937","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"arxiv_version","alias_value":"2411.11937v1","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.11937","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"pith_short_12","alias_value":"NDMPVL2LZO4L","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"pith_short_16","alias_value":"NDMPVL2LZO4LCFDX","created_at":"2026-07-05T09:37:14Z"},{"alias_kind":"pith_short_8","alias_value":"NDMPVL2L","created_at":"2026-07-05T09:37:14Z"}],"graph_snapshots":[{"event_id":"sha256:a0a80e4330debbb48606b8503b48b9b1071368141cf0f84d348b20038e3915b1","target":"graph","created_at":"2026-07-05T09:37:14Z","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/2411.11937/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLMs are increasingly fine-tuned using RLHF datasets to align them with human preferences and values. However, very limited research has investigated which specific human values are operationalized through these datasets. In this paper, we introduce Value Imprint, a framework for auditing and classifying the human values embedded within RLHF datasets. To investigate the viability of this framework, we conducted three case study experiments by auditing the Anthropic/hh-rlhf, OpenAI WebGPT Comparisons, and Alpaca GPT-4-LLM datasets to examine the human values embedded within them. Our analysis i","authors_text":"Aaron Basiletti, Ike Obi, Maham Ghazanfar, Rohan Pant, Srishti Shekhar Agrawal","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-18T16:12:24Z","title":"Value Imprint: A Technique for Auditing the Human Values Embedded in RLHF Datasets"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.11937","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:a0a2d839c9f994048e341974984c59ce368e7d11f3b0431dc0d2247ad7b33ac7","target":"record","created_at":"2026-07-05T09:37:14Z","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":"da4500aeb5d01374b708685c0e8b79c6a1fccd65d3b30682caa66a4b32e1c58a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-18T16:12:24Z","title_canon_sha256":"7c566c6c5dc2d32c21eb921540ccc70bf2fe9af35d1b067377399618953eeee9"},"schema_version":"1.0","source":{"id":"2411.11937","kind":"arxiv","version":1}},"canonical_sha256":"68d8faaf4bcbb8b114779945792feeb5c0dc038b3ad7777674d7ae4a4be53f9f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"68d8faaf4bcbb8b114779945792feeb5c0dc038b3ad7777674d7ae4a4be53f9f","first_computed_at":"2026-07-05T09:37:14.139341Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:37:14.139341Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GwlEkI/H0gy/LvWro7eeIE1mE7bK36cRolGgA7R+K2w7ABMOhDv7wRKI3D0462fTIWw71Awtv9diLi4gQcXOCA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:37:14.139961Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.11937","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a0a2d839c9f994048e341974984c59ce368e7d11f3b0431dc0d2247ad7b33ac7","sha256:a0a80e4330debbb48606b8503b48b9b1071368141cf0f84d348b20038e3915b1"],"state_sha256":"85b91895bd5ae2691ed52cdd91de22e063dad4de7fe5e9742ca41b8367500909"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p5ReLinFTZ7K//5QmyuXsNBHTckmoI5PfR53V5dRLeR+KLMeitHauYtlv/BtpSYvVJ8iXRa/rPT5EEKZZVLqBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T16:57:46.509400Z","bundle_sha256":"8b3d7dcb68c122ebe17c8bf4e8657a9513e9c567ceca330589d2b8715d5ce62e"}}