{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JOW6UEEPZBN6XFAMOMV74QGOSP","short_pith_number":"pith:JOW6UEEP","schema_version":"1.0","canonical_sha256":"4badea108fc85beb940c732bfe40ce93e282e52b4c0a76658250f85383e05e81","source":{"kind":"arxiv","id":"2404.02655","version":2},"attestation_state":"computed","paper":{"title":"Calibrating the Confidence of Large Language Models by Eliciting Fidelity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chong Peng, Linsen Guo, Mianqiu Huang, Mozhi Zhang, Peng Yan, Rundong Shi, Xipeng Qiu, Yaqian Zhou","submitted_at":"2024-04-03T11:36:12Z","abstract_excerpt":"Large language models optimized with techniques like RLHF have achieved good alignment in being helpful and harmless. However, post-alignment, these language models often exhibit overconfidence, where the expressed confidence does not accurately calibrate with their correctness rate. In this paper, we decompose the language model confidence into the \\textit{Uncertainty} about the question and the \\textit{Fidelity} to the answer generated by language models. Then, we propose a plug-and-play method to estimate the confidence of language models. Our method has shown good calibration performance b"},"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":"2404.02655","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-03T11:36:12Z","cross_cats_sorted":[],"title_canon_sha256":"30c0e22e49a87d34d9d119ba7d16853f1d110cf88d74dd1b990d90ea03ed71a3","abstract_canon_sha256":"0adde1da5590b94cb472222805c76d5d850cd31db3f3c7fc621f7d0ff62e7c5e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:17:39.501133Z","signature_b64":"k6Hgh1eZlrg6fbUXFWPFp2ek1+RxDExi8UjOGLnD2OmX78xAqgWx6+UyQ/uUd3FJDX7uFTwvYE4/GM/VjbVxDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4badea108fc85beb940c732bfe40ce93e282e52b4c0a76658250f85383e05e81","last_reissued_at":"2026-07-05T09:17:39.500658Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:17:39.500658Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Calibrating the Confidence of Large Language Models by Eliciting Fidelity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chong Peng, Linsen Guo, Mianqiu Huang, Mozhi Zhang, Peng Yan, Rundong Shi, Xipeng Qiu, Yaqian Zhou","submitted_at":"2024-04-03T11:36:12Z","abstract_excerpt":"Large language models optimized with techniques like RLHF have achieved good alignment in being helpful and harmless. However, post-alignment, these language models often exhibit overconfidence, where the expressed confidence does not accurately calibrate with their correctness rate. In this paper, we decompose the language model confidence into the \\textit{Uncertainty} about the question and the \\textit{Fidelity} to the answer generated by language models. Then, we propose a plug-and-play method to estimate the confidence of language models. Our method has shown good calibration performance b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.02655","kind":"arxiv","version":2},"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/2404.02655/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":"2404.02655","created_at":"2026-07-05T09:17:39.500715+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.02655v2","created_at":"2026-07-05T09:17:39.500715+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.02655","created_at":"2026-07-05T09:17:39.500715+00:00"},{"alias_kind":"pith_short_12","alias_value":"JOW6UEEPZBN6","created_at":"2026-07-05T09:17:39.500715+00:00"},{"alias_kind":"pith_short_16","alias_value":"JOW6UEEPZBN6XFAM","created_at":"2026-07-05T09:17:39.500715+00:00"},{"alias_kind":"pith_short_8","alias_value":"JOW6UEEP","created_at":"2026-07-05T09:17:39.500715+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP","json":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP.json","graph_json":"https://pith.science/api/pith-number/JOW6UEEPZBN6XFAMOMV74QGOSP/graph.json","events_json":"https://pith.science/api/pith-number/JOW6UEEPZBN6XFAMOMV74QGOSP/events.json","paper":"https://pith.science/paper/JOW6UEEP"},"agent_actions":{"view_html":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP","download_json":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP.json","view_paper":"https://pith.science/paper/JOW6UEEP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.02655&json=true","fetch_graph":"https://pith.science/api/pith-number/JOW6UEEPZBN6XFAMOMV74QGOSP/graph.json","fetch_events":"https://pith.science/api/pith-number/JOW6UEEPZBN6XFAMOMV74QGOSP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP/action/storage_attestation","attest_author":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP/action/author_attestation","sign_citation":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP/action/citation_signature","submit_replication":"https://pith.science/pith/JOW6UEEPZBN6XFAMOMV74QGOSP/action/replication_record"}},"created_at":"2026-07-05T09:17:39.500715+00:00","updated_at":"2026-07-05T09:17:39.500715+00:00"}