{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VAHMOTCPY66FMW5WUVSTCTD5RL","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"65626616c33ca14f8b4d6c9bbdc88ae8c024e8eef61d03891061fe532aa71922","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-23T22:22:00Z","title_canon_sha256":"2fdce246582151c38a6ac17d97009b593dfa0734e7c91dda9ff3d80a9c733ba7"},"schema_version":"1.0","source":{"id":"2605.24756","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24756","created_at":"2026-05-26T01:03:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24756v1","created_at":"2026-05-26T01:03:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24756","created_at":"2026-05-26T01:03:56Z"},{"alias_kind":"pith_short_12","alias_value":"VAHMOTCPY66F","created_at":"2026-05-26T01:03:56Z"},{"alias_kind":"pith_short_16","alias_value":"VAHMOTCPY66FMW5W","created_at":"2026-05-26T01:03:56Z"},{"alias_kind":"pith_short_8","alias_value":"VAHMOTCP","created_at":"2026-05-26T01:03:56Z"}],"graph_snapshots":[{"event_id":"sha256:12fd86df44889a821c155df8dac0df6c01d30cd4f0029cb2cf14d23699795a7e","target":"graph","created_at":"2026-05-26T01:03:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.24756/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Satwik Pandey, Shashwat Pandey, Suresh Raghu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-23T22:22:00Z","title":"Proper Scoring Rules for Agentic Uncertainty Quantification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24756","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e4d68e4b53ff28a8e764b24f71b290e43d4a3ed9e1f4e4f17e84cd54153486c6","target":"record","created_at":"2026-05-26T01:03:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"65626616c33ca14f8b4d6c9bbdc88ae8c024e8eef61d03891061fe532aa71922","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-23T22:22:00Z","title_canon_sha256":"2fdce246582151c38a6ac17d97009b593dfa0734e7c91dda9ff3d80a9c733ba7"},"schema_version":"1.0","source":{"id":"2605.24756","kind":"arxiv","version":1}},"canonical_sha256":"a80ec74c4fc7bc565bb6a565314c7d8af827cd813e279c776f39f6f4f091d9fc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a80ec74c4fc7bc565bb6a565314c7d8af827cd813e279c776f39f6f4f091d9fc","first_computed_at":"2026-05-26T01:03:56.615580Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T01:03:56.615580Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VV7pfLwEsHM5z/B4tLMAlAz/nNYm2vO1WLF18sChta9fLKxAYakr0+Yire7pBsCmWiGHajWCqjxdhHpqGW7gCA==","signature_status":"signed_v1","signed_at":"2026-05-26T01:03:56.616383Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.24756","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e4d68e4b53ff28a8e764b24f71b290e43d4a3ed9e1f4e4f17e84cd54153486c6","sha256:12fd86df44889a821c155df8dac0df6c01d30cd4f0029cb2cf14d23699795a7e"],"state_sha256":"99095f3ba0b0c241a23d591defaefde622f97eaa25e3554b6b5f3bf46f0320ab"}