{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YXMEXGXS52RTL2O3ANDRCXKMYV","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":"5d0ca1e92368399fe3a3a9db50d19167579f08595530f4cb13d0a0f972a7f710","cross_cats_sorted":["cs.AI","cs.CL","cs.CY"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-19T02:34:12Z","title_canon_sha256":"770af4933170c715951f3c1553d8b7483218c583e0acdb9fd0fed96a6d68e247"},"schema_version":"1.0","source":{"id":"2311.11202","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.11202","created_at":"2026-07-05T07:59:57Z"},{"alias_kind":"arxiv_version","alias_value":"2311.11202v2","created_at":"2026-07-05T07:59:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.11202","created_at":"2026-07-05T07:59:57Z"},{"alias_kind":"pith_short_12","alias_value":"YXMEXGXS52RT","created_at":"2026-07-05T07:59:57Z"},{"alias_kind":"pith_short_16","alias_value":"YXMEXGXS52RTL2O3","created_at":"2026-07-05T07:59:57Z"},{"alias_kind":"pith_short_8","alias_value":"YXMEXGXS","created_at":"2026-07-05T07:59:57Z"}],"graph_snapshots":[{"event_id":"sha256:516e816fd7699311a69e9da5bf94759c832fb52620b71d868197ddf0176b9900","target":"graph","created_at":"2026-07-05T07:59:57Z","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/2311.11202/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models have shown promise in various tasks but can be affected by undesired data during training, fine-tuning, or alignment. For example, if some unsafe conversations are wrongly annotated as safe ones, the model fine-tuned on these samples may be harmful. Therefore, the correctness of annotations, i.e., the credibility of the dataset, is important. This study focuses on the credibility of real-world datasets, including the popular benchmarks Jigsaw Civil Comments, Anthropic Harmless & Red Team, PKU BeaverTails & SafeRLHF, that can be used for training a harmless language model. Given","authors_text":"Hao Cheng, Jialu Wang, Yang Liu, Zhaowei Zhu","cross_cats":["cs.AI","cs.CL","cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-19T02:34:12Z","title":"Unmasking and Improving Data Credibility: A Study with Datasets for Training Harmless Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.11202","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":"sha256:0ccaaf312966d96194ee0eb75d196996c3b4c0a7692aa55e6f70aa05993541b8","target":"record","created_at":"2026-07-05T07:59:57Z","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":"5d0ca1e92368399fe3a3a9db50d19167579f08595530f4cb13d0a0f972a7f710","cross_cats_sorted":["cs.AI","cs.CL","cs.CY"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-19T02:34:12Z","title_canon_sha256":"770af4933170c715951f3c1553d8b7483218c583e0acdb9fd0fed96a6d68e247"},"schema_version":"1.0","source":{"id":"2311.11202","kind":"arxiv","version":2}},"canonical_sha256":"c5d84b9af2eea335e9db0347115d4cc57b95e0d6cf1c8ea4ce95ca2b4887dbd5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c5d84b9af2eea335e9db0347115d4cc57b95e0d6cf1c8ea4ce95ca2b4887dbd5","first_computed_at":"2026-07-05T07:59:57.440330Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:59:57.440330Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4RUxwEaNT3uAwuNmPAshM160AO2dPl/ioe1vPnYIOILoEos53obxxYFC1ym+dIsI6g38iq6przG/f13S8hJaCg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:59:57.440790Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.11202","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0ccaaf312966d96194ee0eb75d196996c3b4c0a7692aa55e6f70aa05993541b8","sha256:516e816fd7699311a69e9da5bf94759c832fb52620b71d868197ddf0176b9900"],"state_sha256":"e73bfe5f4094aa6cdfa9a2700a282efd0788467ee6c4a2cc6ad90261a1b1d99d"}