{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CDFZ46NYNKIBML2V2IPFLXQWAW","short_pith_number":"pith:CDFZ46NY","schema_version":"1.0","canonical_sha256":"10cb9e79b86a90162f55d21e55de160585b0c2ea4a22be6cae77f999186f96eb","source":{"kind":"arxiv","id":"2405.16856","version":1},"attestation_state":"computed","paper":{"title":"Can We Trust LLMs? Mitigate Overconfidence Bias in LLMs through Knowledge Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hanyuan Zhang, Haoyan Yang, Xingyin Xu, Yirong Bian, Yixuan Wang","submitted_at":"2024-05-27T06:06:36Z","abstract_excerpt":"The study explores mitigating overconfidence bias in LLMs to improve their reliability. We introduce a knowledge transfer (KT) method utilizing chain of thoughts, where \"big\" LLMs impart knowledge to \"small\" LLMs via detailed, sequential reasoning paths. This method uses advanced reasoning of larger models to fine-tune smaller models, enabling them to produce more accurate predictions with calibrated confidence. Experimental evaluation using multiple-choice questions and sentiment analysis across diverse datasets demonstrated the KT method's superiority over the vanilla and question-answer pai"},"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":"2405.16856","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-27T06:06:36Z","cross_cats_sorted":[],"title_canon_sha256":"fde69ce2385fd0caf7ff8bad4ee7150277cf61fbbe00ab15e8ae8759a379ab0f","abstract_canon_sha256":"a66e43e6873c513096623d1db765fe9fe2cca05204b83e633ff94eee6b33c5a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:23:36.583964Z","signature_b64":"1crWVgSRYnqSmtWzNeStGQOF1ZKFr61KymamsTSowG3b9L2ECbrY4Atm4dhhCzB6CXqDuPvJySi6KDJuHr4mCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"10cb9e79b86a90162f55d21e55de160585b0c2ea4a22be6cae77f999186f96eb","last_reissued_at":"2026-07-05T08:23:36.583486Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:23:36.583486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can We Trust LLMs? Mitigate Overconfidence Bias in LLMs through Knowledge Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hanyuan Zhang, Haoyan Yang, Xingyin Xu, Yirong Bian, Yixuan Wang","submitted_at":"2024-05-27T06:06:36Z","abstract_excerpt":"The study explores mitigating overconfidence bias in LLMs to improve their reliability. We introduce a knowledge transfer (KT) method utilizing chain of thoughts, where \"big\" LLMs impart knowledge to \"small\" LLMs via detailed, sequential reasoning paths. This method uses advanced reasoning of larger models to fine-tune smaller models, enabling them to produce more accurate predictions with calibrated confidence. Experimental evaluation using multiple-choice questions and sentiment analysis across diverse datasets demonstrated the KT method's superiority over the vanilla and question-answer pai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.16856","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/2405.16856/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":"2405.16856","created_at":"2026-07-05T08:23:36.583545+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.16856v1","created_at":"2026-07-05T08:23:36.583545+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.16856","created_at":"2026-07-05T08:23:36.583545+00:00"},{"alias_kind":"pith_short_12","alias_value":"CDFZ46NYNKIB","created_at":"2026-07-05T08:23:36.583545+00:00"},{"alias_kind":"pith_short_16","alias_value":"CDFZ46NYNKIBML2V","created_at":"2026-07-05T08:23:36.583545+00:00"},{"alias_kind":"pith_short_8","alias_value":"CDFZ46NY","created_at":"2026-07-05T08:23:36.583545+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2410.13181","citing_title":"AdaSwitch: Adaptive Switching between Small and Large Agents for Effective Cloud-Local Collaborative Learning","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2412.02904","citing_title":"Enhancing Trust in Large Language Models via Uncertainty-Calibrated Fine-Tuning","ref_index":64,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW","json":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW.json","graph_json":"https://pith.science/api/pith-number/CDFZ46NYNKIBML2V2IPFLXQWAW/graph.json","events_json":"https://pith.science/api/pith-number/CDFZ46NYNKIBML2V2IPFLXQWAW/events.json","paper":"https://pith.science/paper/CDFZ46NY"},"agent_actions":{"view_html":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW","download_json":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW.json","view_paper":"https://pith.science/paper/CDFZ46NY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.16856&json=true","fetch_graph":"https://pith.science/api/pith-number/CDFZ46NYNKIBML2V2IPFLXQWAW/graph.json","fetch_events":"https://pith.science/api/pith-number/CDFZ46NYNKIBML2V2IPFLXQWAW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW/action/storage_attestation","attest_author":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW/action/author_attestation","sign_citation":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW/action/citation_signature","submit_replication":"https://pith.science/pith/CDFZ46NYNKIBML2V2IPFLXQWAW/action/replication_record"}},"created_at":"2026-07-05T08:23:36.583545+00:00","updated_at":"2026-07-05T08:23:36.583545+00:00"}