{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:C3I2VXYRTWOWBPS2ODFRCQP7UJ","short_pith_number":"pith:C3I2VXYR","canonical_record":{"source":{"id":"2408.10718","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-08-20T10:40:35Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"07bd5721fa8af85663e50fb416deaf45a47adb5bfb98c27229316c11a4511dce","abstract_canon_sha256":"8b29d5ba6e926ed1dfe5b7a3a384160c043189811ed3f345a8d17724c00dd66b"},"schema_version":"1.0"},"canonical_sha256":"16d1aadf119d9d60be5a70cb1141ffa2465795b9fc5864870f6bb7c9b4e6eccc","source":{"kind":"arxiv","id":"2408.10718","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"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:C3I2VXYRTWOWBPS2ODFRCQP7UJ","target":"record","payload":{"canonical_record":{"source":{"id":"2408.10718","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-08-20T10:40:35Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"07bd5721fa8af85663e50fb416deaf45a47adb5bfb98c27229316c11a4511dce","abstract_canon_sha256":"8b29d5ba6e926ed1dfe5b7a3a384160c043189811ed3f345a8d17724c00dd66b"},"schema_version":"1.0"},"canonical_sha256":"16d1aadf119d9d60be5a70cb1141ffa2465795b9fc5864870f6bb7c9b4e6eccc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:06:43.588950Z","signature_b64":"FT6r3u6nyzgd8uPe5Vvpf9nDa56FEmoYa0WtDr2fKk6oJIX02i+tSMNiPLaxKMrNS8caiG8REbg437W7/VTHCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"16d1aadf119d9d60be5a70cb1141ffa2465795b9fc5864870f6bb7c9b4e6eccc","last_reissued_at":"2026-07-05T09:06:43.588481Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:06:43.588481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2408.10718","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:06:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F34X/qqLzJ3gcvhTw3BWxuOPbbOjvkDPYRUmw0oOvn4HXnNTpWL9y42gIm0yOZ126YtNOD1YsRrHebIdw8JsCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T15:51:01.138021Z"},"content_sha256":"2b2d55fe27cb8841dbc06a333d8aefa7e7a75a1650b19f70c13a7d824be30556","schema_version":"1.0","event_id":"sha256:2b2d55fe27cb8841dbc06a333d8aefa7e7a75a1650b19f70c13a7d824be30556"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:C3I2VXYRTWOWBPS2ODFRCQP7UJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CodeJudge-Eval: Can Large Language Models be Good Judges in Code Understanding?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Annan Li, Hongzhan Lin, Jing Ma, Weixiang Yan, Yuchen Tian, Yuwei Zhao, Ziyang Luo","submitted_at":"2024-08-20T10:40:35Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.10718","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2408.10718/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:06:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E7gvSQCWQunxXV3ZJznflkcLRqXiFLpymGMg4a9qMPWeUrSAEkbAw0s1eOXJUwUsPe3qUZxPeUaSIxfaOcjtDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T15:51:01.138836Z"},"content_sha256":"fc31c4a3baccc9926358d7e8ffe21c9bb5b4395af53713abd0e5b120770b81ec","schema_version":"1.0","event_id":"sha256:fc31c4a3baccc9926358d7e8ffe21c9bb5b4395af53713abd0e5b120770b81ec"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/C3I2VXYRTWOWBPS2ODFRCQP7UJ/bundle.json","state_url":"https://pith.science/pith/C3I2VXYRTWOWBPS2ODFRCQP7UJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/C3I2VXYRTWOWBPS2ODFRCQP7UJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-16T15:51:01Z","links":{"resolver":"https://pith.science/pith/C3I2VXYRTWOWBPS2ODFRCQP7UJ","bundle":"https://pith.science/pith/C3I2VXYRTWOWBPS2ODFRCQP7UJ/bundle.json","state":"https://pith.science/pith/C3I2VXYRTWOWBPS2ODFRCQP7UJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/C3I2VXYRTWOWBPS2ODFRCQP7UJ/bundle.json"},"state":{"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"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dwIMrQ6ta0MYQ7J9etlgx9WDf0hWRUMKYRdcutCpohZ2xz3mTSLhlvR2nWkmWS04KZSd0SF6H8UbWU59GRFMCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T15:51:01.143953Z","bundle_sha256":"3fc9bf5252a61965fb841e7bafc77afe1c6f0dcf04059c9e2d30226cdc11689b"}}