{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AN4IXFBZYKSO7BPDI6YXE3MJNO","short_pith_number":"pith:AN4IXFBZ","schema_version":"1.0","canonical_sha256":"03788b9439c2a4ef85e347b1726d896b99ff56aed9bde8e2dfd607534d53e87c","source":{"kind":"arxiv","id":"2406.05213","version":2},"attestation_state":"computed","paper":{"title":"On Subjective Uncertainty Quantification and Calibration in Natural Language Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Chris Holmes, Ziyu Wang","submitted_at":"2024-06-07T18:54:40Z","abstract_excerpt":"Applications of large language models often involve the generation of free-form responses, in which case uncertainty quantification becomes challenging. This is due to the need to identify task-specific uncertainties (e.g., about the semantics) which appears difficult to define in general cases. This work addresses these challenges from a perspective of Bayesian decision theory, starting from the assumption that our utility is characterized by a similarity measure that compares a generated response with a hypothetical true response. We discuss how this assumption enables principled quantificat"},"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":"2406.05213","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-07T18:54:40Z","cross_cats_sorted":["cs.AI","cs.LG","stat.ML"],"title_canon_sha256":"e444fb199759d29c895ed9f5b83da275656018c1b149d5c7a3b46eb9fbbb6e37","abstract_canon_sha256":"68b059f52c563e28d55ceb4c52922ca4adf3ab7d0b24f2354ca9317e8c35dae8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:22:19.035923Z","signature_b64":"BZHmefWsxjwePc451ZSLtML0eBQQ4Hi6mi5/hv5GJL+kGPTm4QVDzUr1M5EhcFXUMOKE4XCxu8RR4AHWVXdtDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03788b9439c2a4ef85e347b1726d896b99ff56aed9bde8e2dfd607534d53e87c","last_reissued_at":"2026-07-05T09:22:19.035452Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:22:19.035452Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Subjective Uncertainty Quantification and Calibration in Natural Language Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Chris Holmes, Ziyu Wang","submitted_at":"2024-06-07T18:54:40Z","abstract_excerpt":"Applications of large language models often involve the generation of free-form responses, in which case uncertainty quantification becomes challenging. This is due to the need to identify task-specific uncertainties (e.g., about the semantics) which appears difficult to define in general cases. This work addresses these challenges from a perspective of Bayesian decision theory, starting from the assumption that our utility is characterized by a similarity measure that compares a generated response with a hypothetical true response. We discuss how this assumption enables principled quantificat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.05213","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/2406.05213/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":"2406.05213","created_at":"2026-07-05T09:22:19.035510+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.05213v2","created_at":"2026-07-05T09:22:19.035510+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.05213","created_at":"2026-07-05T09:22:19.035510+00:00"},{"alias_kind":"pith_short_12","alias_value":"AN4IXFBZYKSO","created_at":"2026-07-05T09:22:19.035510+00:00"},{"alias_kind":"pith_short_16","alias_value":"AN4IXFBZYKSO7BPD","created_at":"2026-07-05T09:22:19.035510+00:00"},{"alias_kind":"pith_short_8","alias_value":"AN4IXFBZ","created_at":"2026-07-05T09:22:19.035510+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17234","citing_title":"Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2506.10060","citing_title":"Textual Bayes: Quantifying Prompt Uncertainty in LLM-Based Systems","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00455","citing_title":"Concentration and Calibration in Predictive Bayesian Inference","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08974","citing_title":"Confident in a Confidence Score: Investigating the Sensitivity of Confidence Scores to Supervised Fine-Tuning","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17480","citing_title":"Trustworthy deep domain adaptation for wearable photoplethysmography signal analysis with decision-theoretic uncertainty quantification","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO","json":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO.json","graph_json":"https://pith.science/api/pith-number/AN4IXFBZYKSO7BPDI6YXE3MJNO/graph.json","events_json":"https://pith.science/api/pith-number/AN4IXFBZYKSO7BPDI6YXE3MJNO/events.json","paper":"https://pith.science/paper/AN4IXFBZ"},"agent_actions":{"view_html":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO","download_json":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO.json","view_paper":"https://pith.science/paper/AN4IXFBZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.05213&json=true","fetch_graph":"https://pith.science/api/pith-number/AN4IXFBZYKSO7BPDI6YXE3MJNO/graph.json","fetch_events":"https://pith.science/api/pith-number/AN4IXFBZYKSO7BPDI6YXE3MJNO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO/action/storage_attestation","attest_author":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO/action/author_attestation","sign_citation":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO/action/citation_signature","submit_replication":"https://pith.science/pith/AN4IXFBZYKSO7BPDI6YXE3MJNO/action/replication_record"}},"created_at":"2026-07-05T09:22:19.035510+00:00","updated_at":"2026-07-05T09:22:19.035510+00:00"}