{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VAHMOTCPY66FMW5WUVSTCTD5RL","short_pith_number":"pith:VAHMOTCP","schema_version":"1.0","canonical_sha256":"a80ec74c4fc7bc565bb6a565314c7d8af827cd813e279c776f39f6f4f091d9fc","source":{"kind":"arxiv","id":"2605.24756","version":1},"attestation_state":"computed","paper":{"title":"Proper Scoring Rules for Agentic Uncertainty Quantification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Satwik Pandey, Shashwat Pandey, Suresh Raghu","submitted_at":"2026-05-23T22:22:00Z","abstract_excerpt":"Language-model agents increasingly emit uncertainty signals throughout a trajectory, but existing agentic UQ evaluations often conflate ranking usefulness with probabilistic truthfulness. AUROC, AUPRC, risk-coverage, Trajectory ECE, and scalarized trajectory scores evaluate discrimination, binwise calibration, or collapsed summaries, but do not strictly elicit the full prefix-conditioned success-probability trace $q_t = P^{\\pi}(Y=1 | H_t)$. Building on prequential proper scoring, we introduce the Trajectory Proper Score (TPS), a predictor-agnostic family of strictly proper trajectory-level sco"},"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":"2605.24756","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-23T22:22:00Z","cross_cats_sorted":[],"title_canon_sha256":"2fdce246582151c38a6ac17d97009b593dfa0734e7c91dda9ff3d80a9c733ba7","abstract_canon_sha256":"65626616c33ca14f8b4d6c9bbdc88ae8c024e8eef61d03891061fe532aa71922"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:56.616383Z","signature_b64":"VV7pfLwEsHM5z/B4tLMAlAz/nNYm2vO1WLF18sChta9fLKxAYakr0+Yire7pBsCmWiGHajWCqjxdhHpqGW7gCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a80ec74c4fc7bc565bb6a565314c7d8af827cd813e279c776f39f6f4f091d9fc","last_reissued_at":"2026-05-26T01:03:56.615580Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:56.615580Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Proper Scoring Rules for Agentic Uncertainty Quantification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Satwik Pandey, Shashwat Pandey, Suresh Raghu","submitted_at":"2026-05-23T22:22:00Z","abstract_excerpt":"Language-model agents increasingly emit uncertainty signals throughout a trajectory, but existing agentic UQ evaluations often conflate ranking usefulness with probabilistic truthfulness. AUROC, AUPRC, risk-coverage, Trajectory ECE, and scalarized trajectory scores evaluate discrimination, binwise calibration, or collapsed summaries, but do not strictly elicit the full prefix-conditioned success-probability trace $q_t = P^{\\pi}(Y=1 | H_t)$. Building on prequential proper scoring, we introduce the Trajectory Proper Score (TPS), a predictor-agnostic family of strictly proper trajectory-level sco"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24756","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/2605.24756/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":"2605.24756","created_at":"2026-05-26T01:03:56.615705+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24756v1","created_at":"2026-05-26T01:03:56.615705+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24756","created_at":"2026-05-26T01:03:56.615705+00:00"},{"alias_kind":"pith_short_12","alias_value":"VAHMOTCPY66F","created_at":"2026-05-26T01:03:56.615705+00:00"},{"alias_kind":"pith_short_16","alias_value":"VAHMOTCPY66FMW5W","created_at":"2026-05-26T01:03:56.615705+00:00"},{"alias_kind":"pith_short_8","alias_value":"VAHMOTCP","created_at":"2026-05-26T01:03:56.615705+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/VAHMOTCPY66FMW5WUVSTCTD5RL","json":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL.json","graph_json":"https://pith.science/api/pith-number/VAHMOTCPY66FMW5WUVSTCTD5RL/graph.json","events_json":"https://pith.science/api/pith-number/VAHMOTCPY66FMW5WUVSTCTD5RL/events.json","paper":"https://pith.science/paper/VAHMOTCP"},"agent_actions":{"view_html":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL","download_json":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL.json","view_paper":"https://pith.science/paper/VAHMOTCP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24756&json=true","fetch_graph":"https://pith.science/api/pith-number/VAHMOTCPY66FMW5WUVSTCTD5RL/graph.json","fetch_events":"https://pith.science/api/pith-number/VAHMOTCPY66FMW5WUVSTCTD5RL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL/action/storage_attestation","attest_author":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL/action/author_attestation","sign_citation":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL/action/citation_signature","submit_replication":"https://pith.science/pith/VAHMOTCPY66FMW5WUVSTCTD5RL/action/replication_record"}},"created_at":"2026-05-26T01:03:56.615705+00:00","updated_at":"2026-05-26T01:03:56.615705+00:00"}