{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3443LEDA5D676W7FUTGTFEJ423","short_pith_number":"pith:3443LEDA","schema_version":"1.0","canonical_sha256":"df39b59060e8fdff5be5a4cd32913cd6cd17bd25711ea80cd05d180f69ba1af4","source":{"kind":"arxiv","id":"2606.24420","version":1},"attestation_state":"computed","paper":{"title":"Beyond Logprobs: A Multi-Signal Confidence Engine for LLM-Based Document Field Extraction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Nitesh Kumar","submitted_at":"2026-06-23T10:58:08Z","abstract_excerpt":"In high-stakes document processing pipelines, including financial reconciliation, compliance verification, and procurement automation, an LLM extraction that is silently wrong is more dangerous than one that is visibly absent. The central challenge is not extraction accuracy alone but reliable confidence estimation: knowing, field by field, whether an extraction can be trusted for automation or deferred to human review. Token-level log-probabilities, verbalized confidence, and multi-sample self-consistency all collapse toward all-positive behaviour at practical thresholds, offering no reliable"},"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":"2606.24420","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-23T10:58:08Z","cross_cats_sorted":[],"title_canon_sha256":"319a736511d37afc58490870908a8e70a8b65b73b36defe01af126287984b3ca","abstract_canon_sha256":"a66be012159f882e9b3b0e05d282ea79a63689534dd788464be2dccc9213995e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:15:29.967180Z","signature_b64":"18UAbpsGXgSRRMxEtA071saPjHbF7hT/Ovq6XoVQ8YkZ2p6JabMEcDDHbI6p7BveLZSDHd1wbu8PgQVwhN5CCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df39b59060e8fdff5be5a4cd32913cd6cd17bd25711ea80cd05d180f69ba1af4","last_reissued_at":"2026-06-24T01:15:29.966825Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:15:29.966825Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Logprobs: A Multi-Signal Confidence Engine for LLM-Based Document Field Extraction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Nitesh Kumar","submitted_at":"2026-06-23T10:58:08Z","abstract_excerpt":"In high-stakes document processing pipelines, including financial reconciliation, compliance verification, and procurement automation, an LLM extraction that is silently wrong is more dangerous than one that is visibly absent. The central challenge is not extraction accuracy alone but reliable confidence estimation: knowing, field by field, whether an extraction can be trusted for automation or deferred to human review. Token-level log-probabilities, verbalized confidence, and multi-sample self-consistency all collapse toward all-positive behaviour at practical thresholds, offering no reliable"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24420","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/2606.24420/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":"2606.24420","created_at":"2026-06-24T01:15:29.966890+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.24420v1","created_at":"2026-06-24T01:15:29.966890+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24420","created_at":"2026-06-24T01:15:29.966890+00:00"},{"alias_kind":"pith_short_12","alias_value":"3443LEDA5D67","created_at":"2026-06-24T01:15:29.966890+00:00"},{"alias_kind":"pith_short_16","alias_value":"3443LEDA5D676W7F","created_at":"2026-06-24T01:15:29.966890+00:00"},{"alias_kind":"pith_short_8","alias_value":"3443LEDA","created_at":"2026-06-24T01:15:29.966890+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423","json":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423.json","graph_json":"https://pith.science/api/pith-number/3443LEDA5D676W7FUTGTFEJ423/graph.json","events_json":"https://pith.science/api/pith-number/3443LEDA5D676W7FUTGTFEJ423/events.json","paper":"https://pith.science/paper/3443LEDA"},"agent_actions":{"view_html":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423","download_json":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423.json","view_paper":"https://pith.science/paper/3443LEDA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.24420&json=true","fetch_graph":"https://pith.science/api/pith-number/3443LEDA5D676W7FUTGTFEJ423/graph.json","fetch_events":"https://pith.science/api/pith-number/3443LEDA5D676W7FUTGTFEJ423/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423/action/storage_attestation","attest_author":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423/action/author_attestation","sign_citation":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423/action/citation_signature","submit_replication":"https://pith.science/pith/3443LEDA5D676W7FUTGTFEJ423/action/replication_record"}},"created_at":"2026-06-24T01:15:29.966890+00:00","updated_at":"2026-06-24T01:15:29.966890+00:00"}