{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EAFIBFKVQU7ZSLMPAR76SWIBU4","short_pith_number":"pith:EAFIBFKV","schema_version":"1.0","canonical_sha256":"200a809555853f992d8f047fe95901a7186c3634de029a53252ae054537b46b5","source":{"kind":"arxiv","id":"2406.12053","version":1},"attestation_state":"computed","paper":{"title":"InternalInspector $I^2$: Robust Confidence Estimation in LLMs through Internal States","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ankith Mohan, Chang-Tien Lu, Jianfeng He, Lifu Huang, Ming Jin, Mohammad Beigi, Qifan Wang, Runing Yang, Ying Shen, Zihao Lin","submitted_at":"2024-06-17T19:46:05Z","abstract_excerpt":"Despite their vast capabilities, Large Language Models (LLMs) often struggle with generating reliable outputs, frequently producing high-confidence inaccuracies known as hallucinations. Addressing this challenge, our research introduces InternalInspector, a novel framework designed to enhance confidence estimation in LLMs by leveraging contrastive learning on internal states including attention states, feed-forward states, and activation states of all layers. Unlike existing methods that primarily focus on the final activation state, InternalInspector conducts a comprehensive analysis across a"},"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":"2406.12053","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T19:46:05Z","cross_cats_sorted":[],"title_canon_sha256":"bcfec54c5d28de3a155a340896741736e4bbda3fbd0029059a8f83abfc030655","abstract_canon_sha256":"a53719d71c75b2d9b68d9ab0b40ff9ba1778f623c1c636a6e8fccbdd4bb720cf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:33:05.839958Z","signature_b64":"2oeFmA7j/Q4UrNQo8yUC8TCROE9i5irE7D++qpn4eM40d74s9aqO5IK7P2wBwY+sW8Q2X6wlLBULxLAJwnqaCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"200a809555853f992d8f047fe95901a7186c3634de029a53252ae054537b46b5","last_reissued_at":"2026-07-05T08:33:05.839490Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:33:05.839490Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"InternalInspector $I^2$: Robust Confidence Estimation in LLMs through Internal States","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ankith Mohan, Chang-Tien Lu, Jianfeng He, Lifu Huang, Ming Jin, Mohammad Beigi, Qifan Wang, Runing Yang, Ying Shen, Zihao Lin","submitted_at":"2024-06-17T19:46:05Z","abstract_excerpt":"Despite their vast capabilities, Large Language Models (LLMs) often struggle with generating reliable outputs, frequently producing high-confidence inaccuracies known as hallucinations. Addressing this challenge, our research introduces InternalInspector, a novel framework designed to enhance confidence estimation in LLMs by leveraging contrastive learning on internal states including attention states, feed-forward states, and activation states of all layers. Unlike existing methods that primarily focus on the final activation state, InternalInspector conducts a comprehensive analysis across a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12053","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/2406.12053/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":"2406.12053","created_at":"2026-07-05T08:33:05.839545+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.12053v1","created_at":"2026-07-05T08:33:05.839545+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12053","created_at":"2026-07-05T08:33:05.839545+00:00"},{"alias_kind":"pith_short_12","alias_value":"EAFIBFKVQU7Z","created_at":"2026-07-05T08:33:05.839545+00:00"},{"alias_kind":"pith_short_16","alias_value":"EAFIBFKVQU7ZSLMP","created_at":"2026-07-05T08:33:05.839545+00:00"},{"alias_kind":"pith_short_8","alias_value":"EAFIBFKV","created_at":"2026-07-05T08:33:05.839545+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.03998","citing_title":"Toward Better Generalisation in Uncertainty Estimators: Leveraging Data-Agnostic Features","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4","json":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4.json","graph_json":"https://pith.science/api/pith-number/EAFIBFKVQU7ZSLMPAR76SWIBU4/graph.json","events_json":"https://pith.science/api/pith-number/EAFIBFKVQU7ZSLMPAR76SWIBU4/events.json","paper":"https://pith.science/paper/EAFIBFKV"},"agent_actions":{"view_html":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4","download_json":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4.json","view_paper":"https://pith.science/paper/EAFIBFKV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.12053&json=true","fetch_graph":"https://pith.science/api/pith-number/EAFIBFKVQU7ZSLMPAR76SWIBU4/graph.json","fetch_events":"https://pith.science/api/pith-number/EAFIBFKVQU7ZSLMPAR76SWIBU4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4/action/storage_attestation","attest_author":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4/action/author_attestation","sign_citation":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4/action/citation_signature","submit_replication":"https://pith.science/pith/EAFIBFKVQU7ZSLMPAR76SWIBU4/action/replication_record"}},"created_at":"2026-07-05T08:33:05.839545+00:00","updated_at":"2026-07-05T08:33:05.839545+00:00"}