{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:U7YYKK4MBBTTNMWWWZLDBKRI3E","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":"aa190b13477e1c13e8fc5905af4c4da24965768a7d948749c4fb819504b6bf53","cross_cats_sorted":["cs.AI","cs.IT","cs.LG","math.IT"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T00:57:51Z","title_canon_sha256":"3fd87d5deb5441d943835b20bfedeb99ac007ae52968b8765859065729483f05"},"schema_version":"1.0","source":{"id":"2605.19228","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.19228","created_at":"2026-05-20T01:05:34Z"},{"alias_kind":"arxiv_version","alias_value":"2605.19228v1","created_at":"2026-05-20T01:05:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19228","created_at":"2026-05-20T01:05:34Z"},{"alias_kind":"pith_short_12","alias_value":"U7YYKK4MBBTT","created_at":"2026-05-20T01:05:34Z"},{"alias_kind":"pith_short_16","alias_value":"U7YYKK4MBBTTNMWW","created_at":"2026-05-20T01:05:34Z"},{"alias_kind":"pith_short_8","alias_value":"U7YYKK4M","created_at":"2026-05-20T01:05:34Z"}],"graph_snapshots":[{"event_id":"sha256:68278fd2bfbfde98aa79f2d310a7d5fb3505085dbf0d60807ca269732f49a47f","target":"graph","created_at":"2026-05-20T01:05:34Z","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/2605.19228/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models have achieved strong performance on reasoning tasks with objective answers by generating step-by-step solutions, but diagnosing where a multi-step reasoning trace might fail remains difficult. Confidence estimation offers a diagnostic signal, yet existing methods are restricted to final answers or require internal model access. In this paper, we introduce Stepwise Confidence Attribution (SCA), a framework for closed-source LLMs that assigns step-level confidence based only on generated reasoning traces. SCA applies the Information Bottleneck principle: steps aligning with","authors_text":"Dengjia Zhang, Hua Wei, Lu Cheng, Tiejin Chen, Xiaoou Liu, Yaqing Wang","cross_cats":["cs.AI","cs.IT","cs.LG","math.IT"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T00:57:51Z","title":"Diagnosing Multi-step Reasoning Failures in Black-box LLMs via Stepwise Confidence Attribution"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19228","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":"sha256:32125bbdf20dd7a12fc15fa1bee35787dad2e46da368e5e64700ffa33acae5f3","target":"record","created_at":"2026-05-20T01:05:34Z","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":"aa190b13477e1c13e8fc5905af4c4da24965768a7d948749c4fb819504b6bf53","cross_cats_sorted":["cs.AI","cs.IT","cs.LG","math.IT"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T00:57:51Z","title_canon_sha256":"3fd87d5deb5441d943835b20bfedeb99ac007ae52968b8765859065729483f05"},"schema_version":"1.0","source":{"id":"2605.19228","kind":"arxiv","version":1}},"canonical_sha256":"a7f1852b8c086736b2d6b65630aa28d9158d9b0433de736da3e885fd9ba08782","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a7f1852b8c086736b2d6b65630aa28d9158d9b0433de736da3e885fd9ba08782","first_computed_at":"2026-05-20T01:05:34.525942Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:05:34.525942Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9iuipZ0H6S8RYGxBqX7NsvLR7jHPyKQVO1Hnl7ubuXS4hqv3KdKPyxEK8ZaGdXgAILoSGheaHA8TyAPCSh02BQ==","signature_status":"signed_v1","signed_at":"2026-05-20T01:05:34.526580Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.19228","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:32125bbdf20dd7a12fc15fa1bee35787dad2e46da368e5e64700ffa33acae5f3","sha256:68278fd2bfbfde98aa79f2d310a7d5fb3505085dbf0d60807ca269732f49a47f"],"state_sha256":"af5262a75d78bcfb2e12f880cf9cf5208a385a0801191b79c22a4bdfaaeda350"}