{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CFPTD5QZD2RLRINWYW4APHX4HU","short_pith_number":"pith:CFPTD5QZ","schema_version":"1.0","canonical_sha256":"115f31f6191ea2b8a1b6c5b8079efc3d3d24a95d643d44d623df8d080444ef0f","source":{"kind":"arxiv","id":"2404.09127","version":3},"attestation_state":"computed","paper":{"title":"Confidence Calibration and Rationalization for LLMs via Multi-Agent Deliberation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Hu, Dheeraj Rajagopal, Dongyeop Kang, Ruixin Yang, Shirley Anugrah Hayati","submitted_at":"2024-04-14T02:40:43Z","abstract_excerpt":"Uncertainty estimation is a significant issue for current large language models (LLMs) that are generally poorly calibrated and over-confident, especially with reinforcement learning from human feedback (RLHF). Unlike humans, whose decisions and confidences not only stem from intrinsic beliefs but can also be adjusted through daily observations, existing calibration methods for LLMs focus on estimating or eliciting individual confidence without taking full advantage of the \"Collective Wisdom\": the interaction among multiple LLMs that can collectively improve both accuracy and calibration. In t"},"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":"2404.09127","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-14T02:40:43Z","cross_cats_sorted":[],"title_canon_sha256":"0f84457d4e11d0dbbb47899ed74acede7873c3767458020c50a8b74814d501b6","abstract_canon_sha256":"f7bf02554fbe8f3b795168399df1dc8d189ea63aac45295e715917e6b0aeea1b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:17:49.278100Z","signature_b64":"LRGGVnlT8QC+DCS4JphEGMtAhKfr2bNKRmqjw6ly1SqKNesQSa+zc1r3YmyA8qH0RXUrPxoIkhv6urClae0yBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"115f31f6191ea2b8a1b6c5b8079efc3d3d24a95d643d44d623df8d080444ef0f","last_reissued_at":"2026-07-05T08:17:49.277593Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:17:49.277593Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Confidence Calibration and Rationalization for LLMs via Multi-Agent Deliberation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Hu, Dheeraj Rajagopal, Dongyeop Kang, Ruixin Yang, Shirley Anugrah Hayati","submitted_at":"2024-04-14T02:40:43Z","abstract_excerpt":"Uncertainty estimation is a significant issue for current large language models (LLMs) that are generally poorly calibrated and over-confident, especially with reinforcement learning from human feedback (RLHF). Unlike humans, whose decisions and confidences not only stem from intrinsic beliefs but can also be adjusted through daily observations, existing calibration methods for LLMs focus on estimating or eliciting individual confidence without taking full advantage of the \"Collective Wisdom\": the interaction among multiple LLMs that can collectively improve both accuracy and calibration. In t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.09127","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/2404.09127/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":"2404.09127","created_at":"2026-07-05T08:17:49.277657+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.09127v3","created_at":"2026-07-05T08:17:49.277657+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.09127","created_at":"2026-07-05T08:17:49.277657+00:00"},{"alias_kind":"pith_short_12","alias_value":"CFPTD5QZD2RL","created_at":"2026-07-05T08:17:49.277657+00:00"},{"alias_kind":"pith_short_16","alias_value":"CFPTD5QZD2RLRINW","created_at":"2026-07-05T08:17:49.277657+00:00"},{"alias_kind":"pith_short_8","alias_value":"CFPTD5QZ","created_at":"2026-07-05T08:17:49.277657+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23414","citing_title":"When Planning Fails Despite Correct Execution: On Epistemic Calibration for LLM-Based Multi-Agent Systems","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU","json":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU.json","graph_json":"https://pith.science/api/pith-number/CFPTD5QZD2RLRINWYW4APHX4HU/graph.json","events_json":"https://pith.science/api/pith-number/CFPTD5QZD2RLRINWYW4APHX4HU/events.json","paper":"https://pith.science/paper/CFPTD5QZ"},"agent_actions":{"view_html":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU","download_json":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU.json","view_paper":"https://pith.science/paper/CFPTD5QZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.09127&json=true","fetch_graph":"https://pith.science/api/pith-number/CFPTD5QZD2RLRINWYW4APHX4HU/graph.json","fetch_events":"https://pith.science/api/pith-number/CFPTD5QZD2RLRINWYW4APHX4HU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU/action/storage_attestation","attest_author":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU/action/author_attestation","sign_citation":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU/action/citation_signature","submit_replication":"https://pith.science/pith/CFPTD5QZD2RLRINWYW4APHX4HU/action/replication_record"}},"created_at":"2026-07-05T08:17:49.277657+00:00","updated_at":"2026-07-05T08:17:49.277657+00:00"}