{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5CTSE3WST6I6DAP4H676JBEBXT","short_pith_number":"pith:5CTSE3WS","canonical_record":{"source":{"id":"2505.08200","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-13T03:30:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f4a647ec692848eb57b756f619a738b56524e15be25819bfdf7a1b9febe8d2a7","abstract_canon_sha256":"c08c96ee020c046b97f314868e07fc3ea70dbcef05afd18d96c97ab97cdd3a33"},"schema_version":"1.0"},"canonical_sha256":"e8a7226ed29f91e181fc3fbfe48481bcdef99ebe328fd70fbd312c2bec4d4f84","source":{"kind":"arxiv","id":"2505.08200","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.08200","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"arxiv_version","alias_value":"2505.08200v1","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08200","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"pith_short_12","alias_value":"5CTSE3WST6I6","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"pith_short_16","alias_value":"5CTSE3WST6I6DAP4","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"pith_short_8","alias_value":"5CTSE3WS","created_at":"2026-07-05T11:02:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5CTSE3WST6I6DAP4H676JBEBXT","target":"record","payload":{"canonical_record":{"source":{"id":"2505.08200","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-13T03:30:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f4a647ec692848eb57b756f619a738b56524e15be25819bfdf7a1b9febe8d2a7","abstract_canon_sha256":"c08c96ee020c046b97f314868e07fc3ea70dbcef05afd18d96c97ab97cdd3a33"},"schema_version":"1.0"},"canonical_sha256":"e8a7226ed29f91e181fc3fbfe48481bcdef99ebe328fd70fbd312c2bec4d4f84","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:26.388775Z","signature_b64":"5r0A7sm9FafCP+ER6rZIE83gOIyvGrO69yAoRVQ0eCMSNXSZLNTHiL6bfeM0RTRsWPcp5W9f5jCiR3Kt9DapCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8a7226ed29f91e181fc3fbfe48481bcdef99ebe328fd70fbd312c2bec4d4f84","last_reissued_at":"2026-07-05T11:02:26.388328Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:26.388328Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.08200","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":"2026-07-05T11:02:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FVctLkWBQx7pts4wJroqmztaPvQk0BMrHS/KU7ebNtSfoIxXmuCmZ7E7Gz5nyJYZj46L+v15jYT7qmPG+4MkCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T05:53:58.185485Z"},"content_sha256":"2976133c8839e774aa79e783f41b98c561e3942d2cb20621d696d47eb5c6643c","schema_version":"1.0","event_id":"sha256:2976133c8839e774aa79e783f41b98c561e3942d2cb20621d696d47eb5c6643c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5CTSE3WST6I6DAP4H676JBEBXT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Head to Predict and a Head to Question: Pre-trained Uncertainty Quantification Heads for Hallucination Detection in LLM Outputs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Akim Tsvigun, Artem Shelmanov, Artem Vazhentsev, Caiqi Zhang, Ekaterina Fadeeva, Igor Kiselev, Ivan Tsvigun, Mrinmaya Sachan, Nico Daheim, Preslav Nakov, Timothy Baldwin, Zhuohan Xie","submitted_at":"2025-05-13T03:30:26Z","abstract_excerpt":"Large Language Models (LLMs) have the tendency to hallucinate, i.e., to sporadically generate false or fabricated information. This presents a major challenge, as hallucinations often appear highly convincing and users generally lack the tools to detect them. Uncertainty quantification (UQ) provides a framework for assessing the reliability of model outputs, aiding in the identification of potential hallucinations. In this work, we introduce pre-trained UQ heads: supervised auxiliary modules for LLMs that substantially enhance their ability to capture uncertainty compared to unsupervised UQ me"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08200","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/2505.08200/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":"2026-07-05T11:02:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"txuGLEhMdSC6C/GC+rWTv6kjjPaXuUd2oRmVRpGFjbvcvdRq0nmO4v6lxJ4DzrS0zTYRkerYQ1pRFBTKIt8aAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T05:53:58.186018Z"},"content_sha256":"a92c5c6a8bbdffd825ededba2ca3b5cd8484b389f1fcde53d6f84adc34438995","schema_version":"1.0","event_id":"sha256:a92c5c6a8bbdffd825ededba2ca3b5cd8484b389f1fcde53d6f84adc34438995"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5CTSE3WST6I6DAP4H676JBEBXT/bundle.json","state_url":"https://pith.science/pith/5CTSE3WST6I6DAP4H676JBEBXT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5CTSE3WST6I6DAP4H676JBEBXT/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":"2026-08-12T05:53:58Z","links":{"resolver":"https://pith.science/pith/5CTSE3WST6I6DAP4H676JBEBXT","bundle":"https://pith.science/pith/5CTSE3WST6I6DAP4H676JBEBXT/bundle.json","state":"https://pith.science/pith/5CTSE3WST6I6DAP4H676JBEBXT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5CTSE3WST6I6DAP4H676JBEBXT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5CTSE3WST6I6DAP4H676JBEBXT","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":"c08c96ee020c046b97f314868e07fc3ea70dbcef05afd18d96c97ab97cdd3a33","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-13T03:30:26Z","title_canon_sha256":"f4a647ec692848eb57b756f619a738b56524e15be25819bfdf7a1b9febe8d2a7"},"schema_version":"1.0","source":{"id":"2505.08200","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.08200","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"arxiv_version","alias_value":"2505.08200v1","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08200","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"pith_short_12","alias_value":"5CTSE3WST6I6","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"pith_short_16","alias_value":"5CTSE3WST6I6DAP4","created_at":"2026-07-05T11:02:26Z"},{"alias_kind":"pith_short_8","alias_value":"5CTSE3WS","created_at":"2026-07-05T11:02:26Z"}],"graph_snapshots":[{"event_id":"sha256:a92c5c6a8bbdffd825ededba2ca3b5cd8484b389f1fcde53d6f84adc34438995","target":"graph","created_at":"2026-07-05T11:02:26Z","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/2505.08200/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have the tendency to hallucinate, i.e., to sporadically generate false or fabricated information. This presents a major challenge, as hallucinations often appear highly convincing and users generally lack the tools to detect them. Uncertainty quantification (UQ) provides a framework for assessing the reliability of model outputs, aiding in the identification of potential hallucinations. In this work, we introduce pre-trained UQ heads: supervised auxiliary modules for LLMs that substantially enhance their ability to capture uncertainty compared to unsupervised UQ me","authors_text":"Akim Tsvigun, Artem Shelmanov, Artem Vazhentsev, Caiqi Zhang, Ekaterina Fadeeva, Igor Kiselev, Ivan Tsvigun, Mrinmaya Sachan, Nico Daheim, Preslav Nakov, Timothy Baldwin, Zhuohan Xie","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-13T03:30:26Z","title":"A Head to Predict and a Head to Question: Pre-trained Uncertainty Quantification Heads for Hallucination Detection in LLM Outputs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08200","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":"sha256:2976133c8839e774aa79e783f41b98c561e3942d2cb20621d696d47eb5c6643c","target":"record","created_at":"2026-07-05T11:02:26Z","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":"c08c96ee020c046b97f314868e07fc3ea70dbcef05afd18d96c97ab97cdd3a33","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-13T03:30:26Z","title_canon_sha256":"f4a647ec692848eb57b756f619a738b56524e15be25819bfdf7a1b9febe8d2a7"},"schema_version":"1.0","source":{"id":"2505.08200","kind":"arxiv","version":1}},"canonical_sha256":"e8a7226ed29f91e181fc3fbfe48481bcdef99ebe328fd70fbd312c2bec4d4f84","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e8a7226ed29f91e181fc3fbfe48481bcdef99ebe328fd70fbd312c2bec4d4f84","first_computed_at":"2026-07-05T11:02:26.388328Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:02:26.388328Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5r0A7sm9FafCP+ER6rZIE83gOIyvGrO69yAoRVQ0eCMSNXSZLNTHiL6bfeM0RTRsWPcp5W9f5jCiR3Kt9DapCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:02:26.388775Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.08200","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2976133c8839e774aa79e783f41b98c561e3942d2cb20621d696d47eb5c6643c","sha256:a92c5c6a8bbdffd825ededba2ca3b5cd8484b389f1fcde53d6f84adc34438995"],"state_sha256":"76326814ed892c1de50ad2b923ea9161f2cb15a6914e8e2dd5c6b132917260c6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dVoMbKoZcuSBeqyWtSX1ACtEkrMrdpudqr6UihBD9vmD1hDqG5K6vGS68ySzKMRsbyVLCX//bGz0UP7rCZwpBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T05:53:58.189851Z","bundle_sha256":"51e386b523519d2f835a9ea487f7af2dcbb8ad62f7c85ab43f4b591db3d9e0c0"}}