{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:TRM3KP4NL57CBNSDDKJUC5TEVD","short_pith_number":"pith:TRM3KP4N","canonical_record":{"source":{"id":"2311.08685","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-15T04:22:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"54248f96b87e83f31571c15c7d56c02d2f345a0b07406872980a571b7225cd53","abstract_canon_sha256":"2e82bd6ce887ecbc989803307395f8466b49f7225e563641208572b8ece37970"},"schema_version":"1.0"},"canonical_sha256":"9c59b53f8d5f7e20b6431a93417664a8c4802301af783083cd1f67460230a885","source":{"kind":"arxiv","id":"2311.08685","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.08685","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"arxiv_version","alias_value":"2311.08685v3","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08685","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"pith_short_12","alias_value":"TRM3KP4NL57C","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"pith_short_16","alias_value":"TRM3KP4NL57CBNSD","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"pith_short_8","alias_value":"TRM3KP4N","created_at":"2026-07-05T08:02:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:TRM3KP4NL57CBNSDDKJUC5TEVD","target":"record","payload":{"canonical_record":{"source":{"id":"2311.08685","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-15T04:22:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"54248f96b87e83f31571c15c7d56c02d2f345a0b07406872980a571b7225cd53","abstract_canon_sha256":"2e82bd6ce887ecbc989803307395f8466b49f7225e563641208572b8ece37970"},"schema_version":"1.0"},"canonical_sha256":"9c59b53f8d5f7e20b6431a93417664a8c4802301af783083cd1f67460230a885","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:47.875688Z","signature_b64":"cKm0GsMLiyjdQrgVmf1JHE8vTxPURl7h0WtljnHMAwVQmCSYOfErfWXsJQu4CNwULgn0QjC+FOnYtqaJJ2uYCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c59b53f8d5f7e20b6431a93417664a8c4802301af783083cd1f67460230a885","last_reissued_at":"2026-07-05T08:02:47.875187Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:47.875187Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2311.08685","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":"2026-07-05T08:02:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Bf02SaqhNEKf0R/N7tGuBxQPwiKKKpXo9DTYO+JOkNdB529b9T+rklWUpIeqx+PeWrOCUCou/Rfn3n0+0fiiBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T16:49:46.056049Z"},"content_sha256":"3f9222cd0464ca078eb2da98580c600ee7e89ecb584371048cd150803591e9c6","schema_version":"1.0","event_id":"sha256:3f9222cd0464ca078eb2da98580c600ee7e89ecb584371048cd150803591e9c6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:TRM3KP4NL57CBNSDDKJUC5TEVD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Safer-Instruct: Aligning Language Models with Automated Preference Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jieyu Zhao, Kai Chen, Taiwei Shi","submitted_at":"2023-11-15T04:22:22Z","abstract_excerpt":"Reinforcement learning from human feedback (RLHF) is a vital strategy for enhancing model capability in language models. However, annotating preference data for RLHF is a resource-intensive and creativity-demanding process, while existing automatic generation methods face limitations in data diversity and quality. In response, we present Safer-Instruct, a novel pipeline for automatically constructing large-scale preference data. Our approach leverages reversed instruction tuning, instruction induction, and expert model evaluation to efficiently generate high-quality preference data without hum"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08685","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/2311.08685/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":"2026-07-05T08:02:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dQz5NYZugPE8XdnvqhG8LkY+GAlXkD9IYSQJTkv+8vBxfJsYYzGx+jdYBngWT8+MB4aVd5ckTdt/egYY2rb0AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T16:49:46.056727Z"},"content_sha256":"b4515045d1acc6b98152caed07044aca7d6d34862c914e8bac2a97e197b3db9e","schema_version":"1.0","event_id":"sha256:b4515045d1acc6b98152caed07044aca7d6d34862c914e8bac2a97e197b3db9e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TRM3KP4NL57CBNSDDKJUC5TEVD/bundle.json","state_url":"https://pith.science/pith/TRM3KP4NL57CBNSDDKJUC5TEVD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TRM3KP4NL57CBNSDDKJUC5TEVD/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":"2026-08-10T16:49:46Z","links":{"resolver":"https://pith.science/pith/TRM3KP4NL57CBNSDDKJUC5TEVD","bundle":"https://pith.science/pith/TRM3KP4NL57CBNSDDKJUC5TEVD/bundle.json","state":"https://pith.science/pith/TRM3KP4NL57CBNSDDKJUC5TEVD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TRM3KP4NL57CBNSDDKJUC5TEVD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:TRM3KP4NL57CBNSDDKJUC5TEVD","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":"2e82bd6ce887ecbc989803307395f8466b49f7225e563641208572b8ece37970","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-15T04:22:22Z","title_canon_sha256":"54248f96b87e83f31571c15c7d56c02d2f345a0b07406872980a571b7225cd53"},"schema_version":"1.0","source":{"id":"2311.08685","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.08685","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"arxiv_version","alias_value":"2311.08685v3","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08685","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"pith_short_12","alias_value":"TRM3KP4NL57C","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"pith_short_16","alias_value":"TRM3KP4NL57CBNSD","created_at":"2026-07-05T08:02:47Z"},{"alias_kind":"pith_short_8","alias_value":"TRM3KP4N","created_at":"2026-07-05T08:02:47Z"}],"graph_snapshots":[{"event_id":"sha256:b4515045d1acc6b98152caed07044aca7d6d34862c914e8bac2a97e197b3db9e","target":"graph","created_at":"2026-07-05T08:02:47Z","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/2311.08685/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 vital strategy for enhancing model capability in language models. However, annotating preference data for RLHF is a resource-intensive and creativity-demanding process, while existing automatic generation methods face limitations in data diversity and quality. In response, we present Safer-Instruct, a novel pipeline for automatically constructing large-scale preference data. Our approach leverages reversed instruction tuning, instruction induction, and expert model evaluation to efficiently generate high-quality preference data without hum","authors_text":"Jieyu Zhao, Kai Chen, Taiwei Shi","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-15T04:22:22Z","title":"Safer-Instruct: Aligning Language Models with Automated Preference Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08685","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":"sha256:3f9222cd0464ca078eb2da98580c600ee7e89ecb584371048cd150803591e9c6","target":"record","created_at":"2026-07-05T08:02:47Z","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":"2e82bd6ce887ecbc989803307395f8466b49f7225e563641208572b8ece37970","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-15T04:22:22Z","title_canon_sha256":"54248f96b87e83f31571c15c7d56c02d2f345a0b07406872980a571b7225cd53"},"schema_version":"1.0","source":{"id":"2311.08685","kind":"arxiv","version":3}},"canonical_sha256":"9c59b53f8d5f7e20b6431a93417664a8c4802301af783083cd1f67460230a885","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9c59b53f8d5f7e20b6431a93417664a8c4802301af783083cd1f67460230a885","first_computed_at":"2026-07-05T08:02:47.875187Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:02:47.875187Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cKm0GsMLiyjdQrgVmf1JHE8vTxPURl7h0WtljnHMAwVQmCSYOfErfWXsJQu4CNwULgn0QjC+FOnYtqaJJ2uYCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:02:47.875688Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.08685","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3f9222cd0464ca078eb2da98580c600ee7e89ecb584371048cd150803591e9c6","sha256:b4515045d1acc6b98152caed07044aca7d6d34862c914e8bac2a97e197b3db9e"],"state_sha256":"b3013db812e32e5b4cda6cab05efca4be99b2c8842b8aa7f852dad678ca36bd0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OWcohaidaAKXQmUO9GBhtzRHctWx2dTyeAm7h9IyDLcqu8o2Y3FeisJenvq1JFsAI1FbfPAJlQPv8b8Dbz5+AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T16:49:46.062091Z","bundle_sha256":"37c33607493a3e2b40a5fade056725c717aeddfd0aa080b773a795834f5a08c3"}}