{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AUMUXNY4UTKX3HHWD54JPNBMAM","short_pith_number":"pith:AUMUXNY4","schema_version":"1.0","canonical_sha256":"05194bb71ca4d57d9cf61f7897b42c033b5a2e58e8fa3ecb810bbe285b6b388f","source":{"kind":"arxiv","id":"2409.03021","version":1},"attestation_state":"computed","paper":{"title":"CLUE: Concept-Level Uncertainty Estimation for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Andrew Bai, Che-Ping Tsai, Cho-Jui Hsieh, Yu-Hsiang Wang","submitted_at":"2024-09-04T18:27:12Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable proficiency in various natural language generation (NLG) tasks. Previous studies suggest that LLMs' generation process involves uncertainty. However, existing approaches to uncertainty estimation mainly focus on sequence-level uncertainty, overlooking individual pieces of information within sequences. These methods fall short in separately assessing the uncertainty of each component in a sequence. In response, we propose a novel framework for Concept-Level Uncertainty Estimation (CLUE) for LLMs. We leverage LLMs to convert output sequen"},"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":"2409.03021","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-04T18:27:12Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ee813ddfd040db8b2ba58fb05795d438e3acde70ec9a2cb83bbb7a59c20cd610","abstract_canon_sha256":"5d361597aa4b1f2d2446d80f850449dd2df5fb54f64a1209226cf75e1fd9709c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:03:25.125873Z","signature_b64":"Vt5xZgss72MyS7Y8jW981oyZXAVDgdpaL4ZvMNTIrwUT6opF7P572al81s5rQCzMS6L4hS4GVtlQfqhtPY1ECQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"05194bb71ca4d57d9cf61f7897b42c033b5a2e58e8fa3ecb810bbe285b6b388f","last_reissued_at":"2026-07-05T09:03:25.125414Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:03:25.125414Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CLUE: Concept-Level Uncertainty Estimation for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Andrew Bai, Che-Ping Tsai, Cho-Jui Hsieh, Yu-Hsiang Wang","submitted_at":"2024-09-04T18:27:12Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable proficiency in various natural language generation (NLG) tasks. Previous studies suggest that LLMs' generation process involves uncertainty. However, existing approaches to uncertainty estimation mainly focus on sequence-level uncertainty, overlooking individual pieces of information within sequences. These methods fall short in separately assessing the uncertainty of each component in a sequence. In response, we propose a novel framework for Concept-Level Uncertainty Estimation (CLUE) for LLMs. We leverage LLMs to convert output sequen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.03021","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/2409.03021/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":"2409.03021","created_at":"2026-07-05T09:03:25.125469+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.03021v1","created_at":"2026-07-05T09:03:25.125469+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.03021","created_at":"2026-07-05T09:03:25.125469+00:00"},{"alias_kind":"pith_short_12","alias_value":"AUMUXNY4UTKX","created_at":"2026-07-05T09:03:25.125469+00:00"},{"alias_kind":"pith_short_16","alias_value":"AUMUXNY4UTKX3HHW","created_at":"2026-07-05T09:03:25.125469+00:00"},{"alias_kind":"pith_short_8","alias_value":"AUMUXNY4","created_at":"2026-07-05T09:03:25.125469+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05721","citing_title":"SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2604.08974","citing_title":"Confident in a Confidence Score: Investigating the Sensitivity of Confidence Scores to Supervised Fine-Tuning","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM","json":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM.json","graph_json":"https://pith.science/api/pith-number/AUMUXNY4UTKX3HHWD54JPNBMAM/graph.json","events_json":"https://pith.science/api/pith-number/AUMUXNY4UTKX3HHWD54JPNBMAM/events.json","paper":"https://pith.science/paper/AUMUXNY4"},"agent_actions":{"view_html":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM","download_json":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM.json","view_paper":"https://pith.science/paper/AUMUXNY4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.03021&json=true","fetch_graph":"https://pith.science/api/pith-number/AUMUXNY4UTKX3HHWD54JPNBMAM/graph.json","fetch_events":"https://pith.science/api/pith-number/AUMUXNY4UTKX3HHWD54JPNBMAM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM/action/storage_attestation","attest_author":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM/action/author_attestation","sign_citation":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM/action/citation_signature","submit_replication":"https://pith.science/pith/AUMUXNY4UTKX3HHWD54JPNBMAM/action/replication_record"}},"created_at":"2026-07-05T09:03:25.125469+00:00","updated_at":"2026-07-05T09:03:25.125469+00:00"}