{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WSGP3U5BAJEOGJQ3QJSUDEODPW","short_pith_number":"pith:WSGP3U5B","schema_version":"1.0","canonical_sha256":"b48cfdd3a10248e3261b82654191c37da7de91362bbd7e2786776bedfcdab769","source":{"kind":"arxiv","id":"2606.11512","version":1},"attestation_state":"computed","paper":{"title":"SAGE: Answer-Conditioned Uncertainty Targets for Verbal Uncertainty Alignment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Kaiwen Shi, Yanfang Ye, Zheyuan Zhang","submitted_at":"2026-06-09T23:17:16Z","abstract_excerpt":"Large language models increasingly express uncertainty through natural-language statements, yet these expressions often fail to reflect the model's sampled behavior. We study verbal uncertainty alignment as a distributional calibration problem: the appropriate uncertainty target for a prompt should be estimated from repeated model outputs rather than from an isolated response. However, group rollouts alone are insufficient, since the resulting target must provide a useful training signal. Existing targets only partially satisfy this requirement. We propose SAGE, Semantic-Answer Guided Entropy,"},"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":"2606.11512","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T23:17:16Z","cross_cats_sorted":[],"title_canon_sha256":"4167dcf81df67d161117a07365c965db67de44ddb5f716858c9081920aaf6552","abstract_canon_sha256":"62281c9c2882671ed67ac05cc773141cb7b0097e1d443e2d379a07b6e0a7347d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:09:53.404559Z","signature_b64":"Qj0UyRy/C9gu/eXncE2g99PsS6NrJgbc3yx6Zx8ZcoEFOKh4gwB7RDuFLfG32vjyS2V5g1GubqeR7zf9os4dDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b48cfdd3a10248e3261b82654191c37da7de91362bbd7e2786776bedfcdab769","last_reissued_at":"2026-06-11T01:09:53.403671Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:09:53.403671Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SAGE: Answer-Conditioned Uncertainty Targets for Verbal Uncertainty Alignment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Kaiwen Shi, Yanfang Ye, Zheyuan Zhang","submitted_at":"2026-06-09T23:17:16Z","abstract_excerpt":"Large language models increasingly express uncertainty through natural-language statements, yet these expressions often fail to reflect the model's sampled behavior. We study verbal uncertainty alignment as a distributional calibration problem: the appropriate uncertainty target for a prompt should be estimated from repeated model outputs rather than from an isolated response. However, group rollouts alone are insufficient, since the resulting target must provide a useful training signal. Existing targets only partially satisfy this requirement. We propose SAGE, Semantic-Answer Guided Entropy,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11512","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/2606.11512/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":"2606.11512","created_at":"2026-06-11T01:09:53.403815+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.11512v1","created_at":"2026-06-11T01:09:53.403815+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11512","created_at":"2026-06-11T01:09:53.403815+00:00"},{"alias_kind":"pith_short_12","alias_value":"WSGP3U5BAJEO","created_at":"2026-06-11T01:09:53.403815+00:00"},{"alias_kind":"pith_short_16","alias_value":"WSGP3U5BAJEOGJQ3","created_at":"2026-06-11T01:09:53.403815+00:00"},{"alias_kind":"pith_short_8","alias_value":"WSGP3U5B","created_at":"2026-06-11T01:09:53.403815+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/WSGP3U5BAJEOGJQ3QJSUDEODPW","json":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW.json","graph_json":"https://pith.science/api/pith-number/WSGP3U5BAJEOGJQ3QJSUDEODPW/graph.json","events_json":"https://pith.science/api/pith-number/WSGP3U5BAJEOGJQ3QJSUDEODPW/events.json","paper":"https://pith.science/paper/WSGP3U5B"},"agent_actions":{"view_html":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW","download_json":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW.json","view_paper":"https://pith.science/paper/WSGP3U5B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.11512&json=true","fetch_graph":"https://pith.science/api/pith-number/WSGP3U5BAJEOGJQ3QJSUDEODPW/graph.json","fetch_events":"https://pith.science/api/pith-number/WSGP3U5BAJEOGJQ3QJSUDEODPW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW/action/storage_attestation","attest_author":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW/action/author_attestation","sign_citation":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW/action/citation_signature","submit_replication":"https://pith.science/pith/WSGP3U5BAJEOGJQ3QJSUDEODPW/action/replication_record"}},"created_at":"2026-06-11T01:09:53.403815+00:00","updated_at":"2026-06-11T01:09:53.403815+00:00"}