{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GAPDGOPRXPIVL4344XQPSIJFKW","short_pith_number":"pith:GAPDGOPR","schema_version":"1.0","canonical_sha256":"301e3339f1bbd155f37ce5e0f921255595f57ca8858f07b6411111d236345bd4","source":{"kind":"arxiv","id":"2410.14582","version":4},"attestation_state":"computed","paper":{"title":"Do LLMs estimate uncertainty well in instruction-following?","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Christina Heinze-Deml, Jaya Narain, Juyeon Heo, Miao Xiong","submitted_at":"2024-10-18T16:32:10Z","abstract_excerpt":"Large language models (LLMs) could be valuable personal AI agents across various domains, provided they can precisely follow user instructions. However, recent studies have shown significant limitations in LLMs' instruction-following capabilities, raising concerns about their reliability in high-stakes applications. Accurately estimating LLMs' uncertainty in adhering to instructions is critical to mitigating deployment risks. We present, to our knowledge, the first systematic evaluation of the uncertainty estimation abilities of LLMs in the context of instruction-following. Our study identifie"},"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":"2410.14582","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-18T16:32:10Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"d2882bef67be65dd9026d61ec1734966ef5d78679c68a92ab0828ec3972cf066","abstract_canon_sha256":"e208331cb92534fdacef6667f644265852097b96fd02618ffc36cf7813bbb880"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:40:59.097648Z","signature_b64":"cPyKeS23WxFPs0iCGhIYPaGK7bF0TVBXehNMky8RxwXvQcMr+waQdtdZ6tJ09X8KkETSmuevV7D5ldjIlywMCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"301e3339f1bbd155f37ce5e0f921255595f57ca8858f07b6411111d236345bd4","last_reissued_at":"2026-07-05T10:40:59.097128Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:40:59.097128Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do LLMs estimate uncertainty well in instruction-following?","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Christina Heinze-Deml, Jaya Narain, Juyeon Heo, Miao Xiong","submitted_at":"2024-10-18T16:32:10Z","abstract_excerpt":"Large language models (LLMs) could be valuable personal AI agents across various domains, provided they can precisely follow user instructions. However, recent studies have shown significant limitations in LLMs' instruction-following capabilities, raising concerns about their reliability in high-stakes applications. Accurately estimating LLMs' uncertainty in adhering to instructions is critical to mitigating deployment risks. We present, to our knowledge, the first systematic evaluation of the uncertainty estimation abilities of LLMs in the context of instruction-following. Our study identifie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.14582","kind":"arxiv","version":4},"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/2410.14582/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":"2410.14582","created_at":"2026-07-05T10:40:59.097188+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.14582v4","created_at":"2026-07-05T10:40:59.097188+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.14582","created_at":"2026-07-05T10:40:59.097188+00:00"},{"alias_kind":"pith_short_12","alias_value":"GAPDGOPRXPIV","created_at":"2026-07-05T10:40:59.097188+00:00"},{"alias_kind":"pith_short_16","alias_value":"GAPDGOPRXPIVL434","created_at":"2026-07-05T10:40:59.097188+00:00"},{"alias_kind":"pith_short_8","alias_value":"GAPDGOPR","created_at":"2026-07-05T10:40:59.097188+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19509","citing_title":"LLM Doesn't Know What It Doesn't Know: Detecting Epistemic Blind Spots via Cross-Model Attribution Divergence on Clinical Tabular Data","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18117","citing_title":"Online In-Context Distillation for Low-Resource Vision Language Models","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2603.09117","citing_title":"Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11801","citing_title":"CLSGen: A Dual-Head Fine-Tuning Framework for Joint Probabilistic Classification and Verbalized Explanation","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW","json":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW.json","graph_json":"https://pith.science/api/pith-number/GAPDGOPRXPIVL4344XQPSIJFKW/graph.json","events_json":"https://pith.science/api/pith-number/GAPDGOPRXPIVL4344XQPSIJFKW/events.json","paper":"https://pith.science/paper/GAPDGOPR"},"agent_actions":{"view_html":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW","download_json":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW.json","view_paper":"https://pith.science/paper/GAPDGOPR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.14582&json=true","fetch_graph":"https://pith.science/api/pith-number/GAPDGOPRXPIVL4344XQPSIJFKW/graph.json","fetch_events":"https://pith.science/api/pith-number/GAPDGOPRXPIVL4344XQPSIJFKW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW/action/storage_attestation","attest_author":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW/action/author_attestation","sign_citation":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW/action/citation_signature","submit_replication":"https://pith.science/pith/GAPDGOPRXPIVL4344XQPSIJFKW/action/replication_record"}},"created_at":"2026-07-05T10:40:59.097188+00:00","updated_at":"2026-07-05T10:40:59.097188+00:00"}