{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:C3I2VXYRTWOWBPS2ODFRCQP7UJ","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":"8b29d5ba6e926ed1dfe5b7a3a384160c043189811ed3f345a8d17724c00dd66b","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-08-20T10:40:35Z","title_canon_sha256":"07bd5721fa8af85663e50fb416deaf45a47adb5bfb98c27229316c11a4511dce"},"schema_version":"1.0","source":{"id":"2408.10718","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.10718","created_at":"2026-07-05T09:06:43Z"},{"alias_kind":"arxiv_version","alias_value":"2408.10718v2","created_at":"2026-07-05T09:06:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.10718","created_at":"2026-07-05T09:06:43Z"},{"alias_kind":"pith_short_12","alias_value":"C3I2VXYRTWOW","created_at":"2026-07-05T09:06:43Z"},{"alias_kind":"pith_short_16","alias_value":"C3I2VXYRTWOWBPS2","created_at":"2026-07-05T09:06:43Z"},{"alias_kind":"pith_short_8","alias_value":"C3I2VXYR","created_at":"2026-07-05T09:06:43Z"}],"graph_snapshots":[{"event_id":"sha256:fc31c4a3baccc9926358d7e8ffe21c9bb5b4395af53713abd0e5b120770b81ec","target":"graph","created_at":"2026-07-05T09:06:43Z","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/2408.10718/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in large language models (LLMs) have showcased impressive code generation capabilities, primarily evaluated through language-to-code benchmarks. However, these benchmarks may not fully capture a model's code understanding abilities. We introduce CodeJudge-Eval (CJ-Eval), a novel benchmark designed to assess LLMs' code understanding abilities from the perspective of code judging rather than code generation. CJ-Eval challenges models to determine the correctness of provided code solutions, encompassing various error types and compilation issues. By leveraging a diverse set of","authors_text":"Annan Li, Hongzhan Lin, Jing Ma, Weixiang Yan, Yuchen Tian, Yuwei Zhao, Ziyang Luo","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-08-20T10:40:35Z","title":"CodeJudge-Eval: Can Large Language Models be Good Judges in Code Understanding?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.10718","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":"sha256:2b2d55fe27cb8841dbc06a333d8aefa7e7a75a1650b19f70c13a7d824be30556","target":"record","created_at":"2026-07-05T09:06:43Z","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":"8b29d5ba6e926ed1dfe5b7a3a384160c043189811ed3f345a8d17724c00dd66b","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-08-20T10:40:35Z","title_canon_sha256":"07bd5721fa8af85663e50fb416deaf45a47adb5bfb98c27229316c11a4511dce"},"schema_version":"1.0","source":{"id":"2408.10718","kind":"arxiv","version":2}},"canonical_sha256":"16d1aadf119d9d60be5a70cb1141ffa2465795b9fc5864870f6bb7c9b4e6eccc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"16d1aadf119d9d60be5a70cb1141ffa2465795b9fc5864870f6bb7c9b4e6eccc","first_computed_at":"2026-07-05T09:06:43.588481Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:06:43.588481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FT6r3u6nyzgd8uPe5Vvpf9nDa56FEmoYa0WtDr2fKk6oJIX02i+tSMNiPLaxKMrNS8caiG8REbg437W7/VTHCg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:06:43.588950Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.10718","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2b2d55fe27cb8841dbc06a333d8aefa7e7a75a1650b19f70c13a7d824be30556","sha256:fc31c4a3baccc9926358d7e8ffe21c9bb5b4395af53713abd0e5b120770b81ec"],"state_sha256":"8a8b4116c631d83acddeb6cbc02a9d52ee2a87d63c57fcab3e183c7e30cb2850"}