{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KD4JKXZR3UK5UPIX4MMFXWVKT7","short_pith_number":"pith:KD4JKXZR","schema_version":"1.0","canonical_sha256":"50f8955f31dd15da3d17e3185bdaaa9fc92fc3484f239b4f6a7ff7533b00e4ad","source":{"kind":"arxiv","id":"2407.00499","version":3},"attestation_state":"computed","paper":{"title":"ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hengtao Shen, Jinhao Duan, Kaidi Xu, Lu Cheng, Qingni Wang, Xiaofeng Zhu, Xiaoshuang Shi, Yue Zhang, Zhiyuan Wang","submitted_at":"2024-06-29T17:33:07Z","abstract_excerpt":"Uncertainty quantification (UQ) in natural language generation (NLG) tasks remains an open challenge, exacerbated by the closed-source nature of the latest large language models (LLMs). This study investigates applying conformal prediction (CP), which can transform any heuristic uncertainty notion into rigorous prediction sets, to black-box LLMs in open-ended NLG tasks. We introduce a novel uncertainty measure based on self-consistency theory, and then develop a conformal uncertainty criterion by integrating the uncertainty condition aligned with correctness into the CP algorithm. Empirical ev"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2407.00499","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-29T17:33:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"9779b6869dd5464f64030dc1f8fc312a93065cdb734158e041cc6290fb24bc66","abstract_canon_sha256":"196caad848d3b7d115f5499bb14fd3e7be78da88c33008d9e4186b723d987fc0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:36:46.228274Z","signature_b64":"zSB1Z7OV5Qk1hHHSu2C44xIPsZTdZzyfpQ5P114EQ4LHhLGfsgvheu8xiLS7wNHDYXwEaiXg1LgpGdRGMPZhBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50f8955f31dd15da3d17e3185bdaaa9fc92fc3484f239b4f6a7ff7533b00e4ad","last_reissued_at":"2026-07-05T09:36:46.227745Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:36:46.227745Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hengtao Shen, Jinhao Duan, Kaidi Xu, Lu Cheng, Qingni Wang, Xiaofeng Zhu, Xiaoshuang Shi, Yue Zhang, Zhiyuan Wang","submitted_at":"2024-06-29T17:33:07Z","abstract_excerpt":"Uncertainty quantification (UQ) in natural language generation (NLG) tasks remains an open challenge, exacerbated by the closed-source nature of the latest large language models (LLMs). This study investigates applying conformal prediction (CP), which can transform any heuristic uncertainty notion into rigorous prediction sets, to black-box LLMs in open-ended NLG tasks. We introduce a novel uncertainty measure based on self-consistency theory, and then develop a conformal uncertainty criterion by integrating the uncertainty condition aligned with correctness into the CP algorithm. Empirical ev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00499","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/2407.00499/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2407.00499","created_at":"2026-07-05T09:36:46.227803+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.00499v3","created_at":"2026-07-05T09:36:46.227803+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00499","created_at":"2026-07-05T09:36:46.227803+00:00"},{"alias_kind":"pith_short_12","alias_value":"KD4JKXZR3UK5","created_at":"2026-07-05T09:36:46.227803+00:00"},{"alias_kind":"pith_short_16","alias_value":"KD4JKXZR3UK5UPIX","created_at":"2026-07-05T09:36:46.227803+00:00"},{"alias_kind":"pith_short_8","alias_value":"KD4JKXZR","created_at":"2026-07-05T09:36:46.227803+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.05563","citing_title":"A Survey on Uncertainty Quantification of Large Language Models: Taxonomy, Open Research Challenges, and Future Directions","ref_index":219,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7","json":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7.json","graph_json":"https://pith.science/api/pith-number/KD4JKXZR3UK5UPIX4MMFXWVKT7/graph.json","events_json":"https://pith.science/api/pith-number/KD4JKXZR3UK5UPIX4MMFXWVKT7/events.json","paper":"https://pith.science/paper/KD4JKXZR"},"agent_actions":{"view_html":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7","download_json":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7.json","view_paper":"https://pith.science/paper/KD4JKXZR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.00499&json=true","fetch_graph":"https://pith.science/api/pith-number/KD4JKXZR3UK5UPIX4MMFXWVKT7/graph.json","fetch_events":"https://pith.science/api/pith-number/KD4JKXZR3UK5UPIX4MMFXWVKT7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7/action/storage_attestation","attest_author":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7/action/author_attestation","sign_citation":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7/action/citation_signature","submit_replication":"https://pith.science/pith/KD4JKXZR3UK5UPIX4MMFXWVKT7/action/replication_record"}},"created_at":"2026-07-05T09:36:46.227803+00:00","updated_at":"2026-07-05T09:36:46.227803+00:00"}