{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HQKZBTWWG4HABOQTXFUTJ7XYFV","short_pith_number":"pith:HQKZBTWW","schema_version":"1.0","canonical_sha256":"3c1590ced6370e00ba13b96934fef82d41675b835d371de9cc2203d32b6fe15b","source":{"kind":"arxiv","id":"2503.02863","version":2},"attestation_state":"computed","paper":{"title":"SteerConf: Steering LLMs for Confidence Elicitation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Jieming Shi, Qing Li, Tianyuan Jin, Ziang Zhou","submitted_at":"2025-03-04T18:40:49Z","abstract_excerpt":"Large Language Models (LLMs) exhibit impressive performance across diverse domains but often suffer from overconfidence, limiting their reliability in critical applications. We propose SteerConf, a novel framework that systematically steers LLMs' confidence scores to improve their calibration and reliability. SteerConf introduces three key components: (1) a steering prompt strategy that guides LLMs to produce confidence scores in specified directions (e.g., conservative or optimistic) by leveraging prompts with varying steering levels; (2) a steered confidence consistency measure that quantifi"},"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":"2503.02863","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-04T18:40:49Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0dcaa146a9bb6119db676cb976dfeacabbd2cb40788067b6f09db89d303657e4","abstract_canon_sha256":"55c9a9db3fe8ad7057eb44a179c93841e94d431a621c5cd374c4cc15aff3c973"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:40.563540Z","signature_b64":"VO6JsNIwpdFTUDJLfvYZDgK9vcfKhmum2odsdXs8e+dSQY16vKwak1+zDF1FcgI4R30wKLvlPo4Tk0jwRTaGCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c1590ced6370e00ba13b96934fef82d41675b835d371de9cc2203d32b6fe15b","last_reissued_at":"2026-07-05T11:08:40.563062Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:40.563062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SteerConf: Steering LLMs for Confidence Elicitation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Jieming Shi, Qing Li, Tianyuan Jin, Ziang Zhou","submitted_at":"2025-03-04T18:40:49Z","abstract_excerpt":"Large Language Models (LLMs) exhibit impressive performance across diverse domains but often suffer from overconfidence, limiting their reliability in critical applications. We propose SteerConf, a novel framework that systematically steers LLMs' confidence scores to improve their calibration and reliability. SteerConf introduces three key components: (1) a steering prompt strategy that guides LLMs to produce confidence scores in specified directions (e.g., conservative or optimistic) by leveraging prompts with varying steering levels; (2) a steered confidence consistency measure that quantifi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.02863","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/2503.02863/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":"2503.02863","created_at":"2026-07-05T11:08:40.563120+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.02863v2","created_at":"2026-07-05T11:08:40.563120+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.02863","created_at":"2026-07-05T11:08:40.563120+00:00"},{"alias_kind":"pith_short_12","alias_value":"HQKZBTWWG4HA","created_at":"2026-07-05T11:08:40.563120+00:00"},{"alias_kind":"pith_short_16","alias_value":"HQKZBTWWG4HABOQT","created_at":"2026-07-05T11:08:40.563120+00:00"},{"alias_kind":"pith_short_8","alias_value":"HQKZBTWW","created_at":"2026-07-05T11:08:40.563120+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27023","citing_title":"Just how sure are you? Improving Verbalized Uncertainty Calibration in Medical VQA","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19868","citing_title":"A Systematic Evaluation of Black-Box Uncertainty Estimation Methods for Large Language Models","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05799","citing_title":"CaliDist: Calibrating Large Language Models via Behavioral Robustness to Distraction","ref_index":110,"is_internal_anchor":false},{"citing_arxiv_id":"2606.32012","citing_title":"CoMet: Context and Multiplicity Decomposition for Multimodal Uncertainty Estimation","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21652","citing_title":"Look-Closer-Then-Diagnose: Confidence-Aware Ultrasound VQA via Active Zooming","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21652","citing_title":"Look-Closer-Then-Diagnose: Confidence-Aware Ultrasound VQA via Active Zooming","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17187","citing_title":"PluRule: A Benchmark for Moderating Pluralistic Communities on Social Media","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV","json":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV.json","graph_json":"https://pith.science/api/pith-number/HQKZBTWWG4HABOQTXFUTJ7XYFV/graph.json","events_json":"https://pith.science/api/pith-number/HQKZBTWWG4HABOQTXFUTJ7XYFV/events.json","paper":"https://pith.science/paper/HQKZBTWW"},"agent_actions":{"view_html":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV","download_json":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV.json","view_paper":"https://pith.science/paper/HQKZBTWW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.02863&json=true","fetch_graph":"https://pith.science/api/pith-number/HQKZBTWWG4HABOQTXFUTJ7XYFV/graph.json","fetch_events":"https://pith.science/api/pith-number/HQKZBTWWG4HABOQTXFUTJ7XYFV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV/action/storage_attestation","attest_author":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV/action/author_attestation","sign_citation":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV/action/citation_signature","submit_replication":"https://pith.science/pith/HQKZBTWWG4HABOQTXFUTJ7XYFV/action/replication_record"}},"created_at":"2026-07-05T11:08:40.563120+00:00","updated_at":"2026-07-05T11:08:40.563120+00:00"}