{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:2D6EYZLNZV6KEXUSCEEUWEQVFQ","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":"beb579356c933e8af993393c5ee4d3aa722cf83b9f276ba656f71e1c1dfe1790","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T07:13:30Z","title_canon_sha256":"67c63dc3f2e169e4d831c9ce4eca356eadd5bae69a3f3d93cd95a0e76fd3c780"},"schema_version":"1.0","source":{"id":"2409.10044","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.10044","created_at":"2026-07-05T09:54:04Z"},{"alias_kind":"arxiv_version","alias_value":"2409.10044v2","created_at":"2026-07-05T09:54:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.10044","created_at":"2026-07-05T09:54:04Z"},{"alias_kind":"pith_short_12","alias_value":"2D6EYZLNZV6K","created_at":"2026-07-05T09:54:04Z"},{"alias_kind":"pith_short_16","alias_value":"2D6EYZLNZV6KEXUS","created_at":"2026-07-05T09:54:04Z"},{"alias_kind":"pith_short_8","alias_value":"2D6EYZLN","created_at":"2026-07-05T09:54:04Z"}],"graph_snapshots":[{"event_id":"sha256:c8b7447f94948e2faa9857a7150c19d977325465f5e91df9fc5a347f7a3b6292","target":"graph","created_at":"2026-07-05T09:54:04Z","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/2409.10044/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Prompt optimization algorithms for Large Language Models (LLMs) excel in multi-step reasoning but still lack effective uncertainty estimation. This paper introduces a benchmark dataset to evaluate uncertainty metrics, focusing on Answer, Correctness, Aleatoric, and Epistemic Uncertainty. Through analysis of models like GPT-3.5-Turbo and Meta-Llama-3.1-8B-Instruct, we show that current metrics align more with Answer Uncertainty, which reflects output confidence and diversity, rather than Correctness Uncertainty, highlighting the need for improved metrics that are optimization-objective-aware to","authors_text":"Pei-Fu Guo, Shou-De Lin, Yun-Da Tsai","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T07:13:30Z","title":"Benchmarking Large Language Model Uncertainty for Prompt Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.10044","kind":"arxiv","version":2},"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":"sha256:3e5827161e6f66125feae0da54655bbd1752af7b47dcad8a9ea2232f026785f9","target":"record","created_at":"2026-07-05T09:54:04Z","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":"beb579356c933e8af993393c5ee4d3aa722cf83b9f276ba656f71e1c1dfe1790","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T07:13:30Z","title_canon_sha256":"67c63dc3f2e169e4d831c9ce4eca356eadd5bae69a3f3d93cd95a0e76fd3c780"},"schema_version":"1.0","source":{"id":"2409.10044","kind":"arxiv","version":2}},"canonical_sha256":"d0fc4c656dcd7ca25e9211094b12152c15de7f4f9bb2d5612a12053bea2d5b1f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d0fc4c656dcd7ca25e9211094b12152c15de7f4f9bb2d5612a12053bea2d5b1f","first_computed_at":"2026-07-05T09:54:04.406268Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:54:04.406268Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"roumzhDOOZX8X4E91+YMqO8FefwQ8uIApmBTS7RU9fHGMqR4CXPFw69MHYaXIvAmG4i/22BWqcqxDZx7emAMDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:54:04.406713Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.10044","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3e5827161e6f66125feae0da54655bbd1752af7b47dcad8a9ea2232f026785f9","sha256:c8b7447f94948e2faa9857a7150c19d977325465f5e91df9fc5a347f7a3b6292"],"state_sha256":"532e663fe8626102a5fa6a025044ebc05f3cc57cc08cb0f7acdb5f27571be585"}