{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:N7HQU4CCCT7DFYEIQUAYUROKCQ","short_pith_number":"pith:N7HQU4CC","canonical_record":{"source":{"id":"2502.06193","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-10T06:49:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8d987a7b5f29350a0020ad331e0fe7e82d3efa01c01919732da98a05a04eb189","abstract_canon_sha256":"452cf68d1b4bdaf4277ff7d1c621f3faa143192cf5de758629059ee5324d62d2"},"schema_version":"1.0"},"canonical_sha256":"6fcf0a704214fe32e08885018a45ca142d3213909d611aef6bad02f404ae3283","source":{"kind":"arxiv","id":"2502.06193","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06193","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06193v3","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06193","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_12","alias_value":"N7HQU4CCCT7D","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_16","alias_value":"N7HQU4CCCT7DFYEI","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_8","alias_value":"N7HQU4CC","created_at":"2026-07-05T10:51:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:N7HQU4CCCT7DFYEIQUAYUROKCQ","target":"record","payload":{"canonical_record":{"source":{"id":"2502.06193","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-10T06:49:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8d987a7b5f29350a0020ad331e0fe7e82d3efa01c01919732da98a05a04eb189","abstract_canon_sha256":"452cf68d1b4bdaf4277ff7d1c621f3faa143192cf5de758629059ee5324d62d2"},"schema_version":"1.0"},"canonical_sha256":"6fcf0a704214fe32e08885018a45ca142d3213909d611aef6bad02f404ae3283","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:32.497759Z","signature_b64":"bztyYqW4MBothj5gtJ3cyE7MhuHPnfFz+lHiDG3AWB9LjEI1YgyYhYxbIJik/4ffaL1m87Jkxw9AQa1U0oAWAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6fcf0a704214fe32e08885018a45ca142d3213909d611aef6bad02f404ae3283","last_reissued_at":"2026-07-05T10:51:32.497237Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:32.497237Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.06193","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:51:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8s3Tm6ZKDcdOf6pwAnpXFrX4BSytF7MMWM/cSI6JS2iUFD9QoETn3DHuY1boM7oCIVi/hYaq8QXN3CHn7DmbDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:57:20.064716Z"},"content_sha256":"9f0d082decfe8629bbd0cb57ded77097b7acd8bedd9f335439d9af989a277408","schema_version":"1.0","event_id":"sha256:9f0d082decfe8629bbd0cb57ded77097b7acd8bedd9f335439d9af989a277408"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:N7HQU4CCCT7DFYEIQUAYUROKCQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Can LLMs Replace Human Evaluators? An Empirical Study of LLM-as-a-Judge in Software Engineering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Chun Yong Chong, Cuiyun Gao, Guodong Fan, Jiyu Guo, Ruiqi Wang, Xin Xia","submitted_at":"2025-02-10T06:49:29Z","abstract_excerpt":"Recently, large language models (LLMs) have been deployed to tackle various software engineering (SE) tasks like code generation, significantly advancing the automation of SE tasks. However, assessing the quality of these LLM-generated code and text remains challenging. The commonly used Pass@k metric necessitates extensive unit tests and configured environments, demands a high labor cost, and is not suitable for evaluating LLM-generated text. Conventional metrics like BLEU, which measure only lexical rather than semantic similarity, have also come under scrutiny. In response, a new trend has "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06193","kind":"arxiv","version":3},"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/2502.06193/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":"2026-07-05T10:51:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CpV10dSxkheaXIOJXZpAJDNkQao34+irASYFNmipZxJUiFfpFSjaegAVWoFORyhKqI+BvETLZs+sFaC5bKPfAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:57:20.065221Z"},"content_sha256":"0c2ff59fd71892369dff74293bad44118b8bf013a264a8ecc774c98896463b73","schema_version":"1.0","event_id":"sha256:0c2ff59fd71892369dff74293bad44118b8bf013a264a8ecc774c98896463b73"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N7HQU4CCCT7DFYEIQUAYUROKCQ/bundle.json","state_url":"https://pith.science/pith/N7HQU4CCCT7DFYEIQUAYUROKCQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N7HQU4CCCT7DFYEIQUAYUROKCQ/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":"2026-08-09T04:57:20Z","links":{"resolver":"https://pith.science/pith/N7HQU4CCCT7DFYEIQUAYUROKCQ","bundle":"https://pith.science/pith/N7HQU4CCCT7DFYEIQUAYUROKCQ/bundle.json","state":"https://pith.science/pith/N7HQU4CCCT7DFYEIQUAYUROKCQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N7HQU4CCCT7DFYEIQUAYUROKCQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:N7HQU4CCCT7DFYEIQUAYUROKCQ","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":"452cf68d1b4bdaf4277ff7d1c621f3faa143192cf5de758629059ee5324d62d2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-10T06:49:29Z","title_canon_sha256":"8d987a7b5f29350a0020ad331e0fe7e82d3efa01c01919732da98a05a04eb189"},"schema_version":"1.0","source":{"id":"2502.06193","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06193","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06193v3","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06193","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_12","alias_value":"N7HQU4CCCT7D","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_16","alias_value":"N7HQU4CCCT7DFYEI","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_8","alias_value":"N7HQU4CC","created_at":"2026-07-05T10:51:32Z"}],"graph_snapshots":[{"event_id":"sha256:0c2ff59fd71892369dff74293bad44118b8bf013a264a8ecc774c98896463b73","target":"graph","created_at":"2026-07-05T10:51:32Z","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/2502.06193/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recently, large language models (LLMs) have been deployed to tackle various software engineering (SE) tasks like code generation, significantly advancing the automation of SE tasks. However, assessing the quality of these LLM-generated code and text remains challenging. The commonly used Pass@k metric necessitates extensive unit tests and configured environments, demands a high labor cost, and is not suitable for evaluating LLM-generated text. Conventional metrics like BLEU, which measure only lexical rather than semantic similarity, have also come under scrutiny. In response, a new trend has ","authors_text":"Chun Yong Chong, Cuiyun Gao, Guodong Fan, Jiyu Guo, Ruiqi Wang, Xin Xia","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-10T06:49:29Z","title":"Can LLMs Replace Human Evaluators? An Empirical Study of LLM-as-a-Judge in Software Engineering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06193","kind":"arxiv","version":3},"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":"sha256:9f0d082decfe8629bbd0cb57ded77097b7acd8bedd9f335439d9af989a277408","target":"record","created_at":"2026-07-05T10:51:32Z","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":"452cf68d1b4bdaf4277ff7d1c621f3faa143192cf5de758629059ee5324d62d2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-10T06:49:29Z","title_canon_sha256":"8d987a7b5f29350a0020ad331e0fe7e82d3efa01c01919732da98a05a04eb189"},"schema_version":"1.0","source":{"id":"2502.06193","kind":"arxiv","version":3}},"canonical_sha256":"6fcf0a704214fe32e08885018a45ca142d3213909d611aef6bad02f404ae3283","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6fcf0a704214fe32e08885018a45ca142d3213909d611aef6bad02f404ae3283","first_computed_at":"2026-07-05T10:51:32.497237Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:51:32.497237Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bztyYqW4MBothj5gtJ3cyE7MhuHPnfFz+lHiDG3AWB9LjEI1YgyYhYxbIJik/4ffaL1m87Jkxw9AQa1U0oAWAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:51:32.497759Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.06193","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9f0d082decfe8629bbd0cb57ded77097b7acd8bedd9f335439d9af989a277408","sha256:0c2ff59fd71892369dff74293bad44118b8bf013a264a8ecc774c98896463b73"],"state_sha256":"fd7b606701f2ac50c71416de57ec1e69bc1e65ca7af036719b100386fc2c3b15"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HJODgEB1LxP4bzcp4O5AMFqGwP3KFwGplxN9MhyotkC6UBNPVK7HP/iQ79O0amK8f3PG3N871DowZET5xS6qDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T04:57:20.069163Z","bundle_sha256":"1e249363ae371077181ba3dab9a76c68841bfe92eb8649534e445c1a6dc3fb20"}}