{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:R7IFGXYQNHYNXPGY4TOUWELGB7","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":"b247a0ea8460a1432ac444c4cb5f93742535a6bccb9fd5d78b44763e763d88c8","cross_cats_sorted":["cs.AI","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-30T23:03:05Z","title_canon_sha256":"1aca987d93feb8f0626fcd21efcd660d7ebf2d05f33168199c9d21e17b84a030"},"schema_version":"1.0","source":{"id":"2606.00920","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00920","created_at":"2026-06-02T01:04:09Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00920v1","created_at":"2026-06-02T01:04:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00920","created_at":"2026-06-02T01:04:09Z"},{"alias_kind":"pith_short_12","alias_value":"R7IFGXYQNHYN","created_at":"2026-06-02T01:04:09Z"},{"alias_kind":"pith_short_16","alias_value":"R7IFGXYQNHYNXPGY","created_at":"2026-06-02T01:04:09Z"},{"alias_kind":"pith_short_8","alias_value":"R7IFGXYQ","created_at":"2026-06-02T01:04:09Z"}],"graph_snapshots":[{"event_id":"sha256:d87e067c5973e00c3a0a64484b923c6091b4171dfe9906469fedd3bd0940ca84","target":"graph","created_at":"2026-06-02T01:04:09Z","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/2606.00920/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Run-level pass rate overstates retry-free coverage by up to 17.8 percentage points -- and the gap is largest precisely for mid-performing systems. We investigate this accuracy--stability relationship in large language model (LLM) evaluation for deterministic text-conditioned generation, using programming tasks as a concrete testbed. Standard code-generation benchmarks emphasize single-run accuracy or eventual success under repeated sampling, but many deployment settings also require stability: consistent outcomes across repeated invocations under the same task description. We present a repeate","authors_text":"Jiaxi Wen, Lai Yun Choi, Wenbo Ye, Yongxi Zhou","cross_cats":["cs.AI","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-30T23:03:05Z","title":"Accuracy, Stability, and Repeated-Run Reliability of Large Language Models on Deterministic Programming Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00920","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":"sha256:548f8219cf40b9ca6d3dadb496baca83fe7b81941a9f62abd9d3108f6d1b8d88","target":"record","created_at":"2026-06-02T01:04:09Z","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":"b247a0ea8460a1432ac444c4cb5f93742535a6bccb9fd5d78b44763e763d88c8","cross_cats_sorted":["cs.AI","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-30T23:03:05Z","title_canon_sha256":"1aca987d93feb8f0626fcd21efcd660d7ebf2d05f33168199c9d21e17b84a030"},"schema_version":"1.0","source":{"id":"2606.00920","kind":"arxiv","version":1}},"canonical_sha256":"8fd0535f1069f0dbbcd8e4dd4b11660ffaf705cac24e4e0edd5dce2ec4e1fa92","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8fd0535f1069f0dbbcd8e4dd4b11660ffaf705cac24e4e0edd5dce2ec4e1fa92","first_computed_at":"2026-06-02T01:04:09.500139Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:04:09.500139Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7c06LV3RV5IFq6jG0YAidRGZWi4hSceBdWKiXCkNeoAZD/+lKAxYZnj9aFhANrQKhkTzAXBq2EJSOxEyFLllBw==","signature_status":"signed_v1","signed_at":"2026-06-02T01:04:09.500546Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.00920","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:548f8219cf40b9ca6d3dadb496baca83fe7b81941a9f62abd9d3108f6d1b8d88","sha256:d87e067c5973e00c3a0a64484b923c6091b4171dfe9906469fedd3bd0940ca84"],"state_sha256":"75ef83d0667a8d69f9040822e2437f68ec20a13b4d3991e9848c9b366131c74b"}