{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SD3S2CRKHOWGAU5RQ5H4UQDZ5X","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":"5a641f4767405931f65a0b1127fdc70bc5aa1e6c1fac62d03ec01f156b43b3e7","cross_cats_sorted":["cs.AI","cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T13:55:45Z","title_canon_sha256":"bb616aed747aed112c5772d63267bf1b1a1cf48db208d5061fdb2b705f50cef4"},"schema_version":"1.0","source":{"id":"2310.05727","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.05727","created_at":"2026-07-05T06:58:48Z"},{"alias_kind":"arxiv_version","alias_value":"2310.05727v1","created_at":"2026-07-05T06:58:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.05727","created_at":"2026-07-05T06:58:48Z"},{"alias_kind":"pith_short_12","alias_value":"SD3S2CRKHOWG","created_at":"2026-07-05T06:58:48Z"},{"alias_kind":"pith_short_16","alias_value":"SD3S2CRKHOWGAU5R","created_at":"2026-07-05T06:58:48Z"},{"alias_kind":"pith_short_8","alias_value":"SD3S2CRK","created_at":"2026-07-05T06:58:48Z"}],"graph_snapshots":[{"event_id":"sha256:3083120397260c65da32b0974d045687b5fd5fc8525efa51a0f1389e1955c9ef","target":"graph","created_at":"2026-07-05T06:58:48Z","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/2310.05727/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent development of large language models (LLMs) for code like CodeX and CodeT5+ demonstrates tremendous promise in achieving code intelligence. Their ability of synthesizing code that completes a program for performing a pre-defined task has been intensively tested and verified on benchmark datasets including HumanEval and MBPP. Yet, evaluation of these LLMs from more perspectives (than just program synthesis) is also anticipated, considering their broad scope of applications in software engineering. In this paper, we explore the ability of LLMs for testing programs/code. By performing thor","authors_text":"Hao Chen, Weimin Xiong, Yiwen Guo","cross_cats":["cs.AI","cs.LG","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T13:55:45Z","title":"The Program Testing Ability of Large Language Models for Code"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.05727","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":"sha256:4bdf17a4ae10b3dac3d7dfb439c7cdcc2d66a705b84af1d3f0e8bf2179076eb1","target":"record","created_at":"2026-07-05T06:58:48Z","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":"5a641f4767405931f65a0b1127fdc70bc5aa1e6c1fac62d03ec01f156b43b3e7","cross_cats_sorted":["cs.AI","cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T13:55:45Z","title_canon_sha256":"bb616aed747aed112c5772d63267bf1b1a1cf48db208d5061fdb2b705f50cef4"},"schema_version":"1.0","source":{"id":"2310.05727","kind":"arxiv","version":1}},"canonical_sha256":"90f72d0a2a3bac6053b1874fca4079edf6ddb99d9af2197b9f0ff7b7ec8b08f5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"90f72d0a2a3bac6053b1874fca4079edf6ddb99d9af2197b9f0ff7b7ec8b08f5","first_computed_at":"2026-07-05T06:58:48.867374Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:58:48.867374Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dkSsNSSPsXnIkJ3fdNo/SQ+LTq0aF2ecufhamMsH2ijzlRdZFOW6+Xf4s5GE1JvVMUFEH6gOq0NG0yULJPSCAA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:58:48.868055Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.05727","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4bdf17a4ae10b3dac3d7dfb439c7cdcc2d66a705b84af1d3f0e8bf2179076eb1","sha256:3083120397260c65da32b0974d045687b5fd5fc8525efa51a0f1389e1955c9ef"],"state_sha256":"bebb3e09ae759181d0221fb12c7c7002415183f1aba08dbbc0f94010fee9ed56"}