{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:PKRDOS4RULFALVX2JHDQLXLZEX","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":"19b5c3a9b3598e49af6d1677a849745b2327b32d1356fbf1d445f19f7591b8c5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T17:25:59Z","title_canon_sha256":"d5a11805e9f24dc442ccbdc8d84c090ecb608a7349cf15f1c6b501e19a9d52be"},"schema_version":"1.0","source":{"id":"2305.14279","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14279","created_at":"2026-07-05T07:42:39Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14279v4","created_at":"2026-07-05T07:42:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14279","created_at":"2026-07-05T07:42:39Z"},{"alias_kind":"pith_short_12","alias_value":"PKRDOS4RULFA","created_at":"2026-07-05T07:42:39Z"},{"alias_kind":"pith_short_16","alias_value":"PKRDOS4RULFALVX2","created_at":"2026-07-05T07:42:39Z"},{"alias_kind":"pith_short_8","alias_value":"PKRDOS4R","created_at":"2026-07-05T07:42:39Z"}],"graph_snapshots":[{"event_id":"sha256:502ca266bcaba983514d829d467dfd7dd73fa1e528aaf17e2744b244634870ec","target":"graph","created_at":"2026-07-05T07:42:39Z","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/2305.14279/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have achieved widespread success on a variety of in-context few-shot tasks, but this success is typically evaluated via correctness rather than consistency. We argue that self-consistency is an important criteria for valid multi-step reasoning in tasks where the solution is composed of the answers to multiple sub-steps. We propose two types of self-consistency that are particularly important for multi-step reasoning -- hypothetical consistency (a model's ability to predict what its output would be in a hypothetical other context) and compositional consistency (cons","authors_text":"Alicia Parrish, Angelica Chen, Chen Zhao, Jason Phang, Kyunghyun Cho, Samuel R. Bowman, Vishakh Padmakumar","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T17:25:59Z","title":"Two Failures of Self-Consistency in the Multi-Step Reasoning of LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14279","kind":"arxiv","version":4},"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":"sha256:4f6a6ad2557b695b60cc918325f59bab632ec367cbd1b4886a3030232629249b","target":"record","created_at":"2026-07-05T07:42:39Z","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":"19b5c3a9b3598e49af6d1677a849745b2327b32d1356fbf1d445f19f7591b8c5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T17:25:59Z","title_canon_sha256":"d5a11805e9f24dc442ccbdc8d84c090ecb608a7349cf15f1c6b501e19a9d52be"},"schema_version":"1.0","source":{"id":"2305.14279","kind":"arxiv","version":4}},"canonical_sha256":"7aa2374b91a2ca05d6fa49c705dd7925f775feb0a493c90e6e524ae78dbf21af","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7aa2374b91a2ca05d6fa49c705dd7925f775feb0a493c90e6e524ae78dbf21af","first_computed_at":"2026-07-05T07:42:39.818210Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:42:39.818210Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kVLCARKVue+yZBLbrEXUXqJ2g45YgaxMboF5hWDbPRKZ4DH9sLmIq+wvkbvv0t/MhGP09rtN0G6RI6CoTQr4Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:42:39.818617Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.14279","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4f6a6ad2557b695b60cc918325f59bab632ec367cbd1b4886a3030232629249b","sha256:502ca266bcaba983514d829d467dfd7dd73fa1e528aaf17e2744b244634870ec"],"state_sha256":"3bc552cf8f4e63e575153e293e1e540572636b19ef4f1be0e6677adb5022b9d6"}