{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YMVKTNODOKQES4IWFFMY6UWAO6","short_pith_number":"pith:YMVKTNOD","schema_version":"1.0","canonical_sha256":"c32aa9b5c372a049711629598f52c0779452c467d627582e4d96dd8e07b58c3a","source":{"kind":"arxiv","id":"2605.20251","version":1},"attestation_state":"computed","paper":{"title":"ProcBench: Evaluating Process-Level Defects and Control Preservation in LLM Coding Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Chaoyi Xue, Chenbo Liu, Dong Sun, Jiawei He, Jie Jia, Xikai Yang, Yapeng Song","submitted_at":"2026-05-18T08:34:48Z","abstract_excerpt":"Existing benchmarks for LLM coding agents mainly evaluate final outcomes, such as task completion, compilation success, and test pass rates. While these metrics are useful for measuring end-task capability, they provide limited visibility into how an execution unfolds and often miss recurrent process-level failures that arise during multi-step operation. We present ProcBench, a benchmark-oriented framework for evaluating coding-agent trajectories through process defects and control preservation. ProcBench organizes execution failures into a reusable ontology, standardizes heterogeneous logs in"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.20251","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-18T08:34:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1345130dfde1dda18956b8b7d13c6f0c8cd5709fcba7f68762cd8cd08d27b5ae","abstract_canon_sha256":"4277d5d6f288d9f728678d131407444a97fba11c2ee9b3431f893c22dd91e171"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T00:04:22.765992Z","signature_b64":"BGd9ksfb267FJfH6zjMRxK950JIqTnk3tWvymPflUcw9DSb2+9qsG0SPsMCIzuSL5XPvnRnP9TQA3taF5GsJAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c32aa9b5c372a049711629598f52c0779452c467d627582e4d96dd8e07b58c3a","last_reissued_at":"2026-05-21T00:04:22.765496Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T00:04:22.765496Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ProcBench: Evaluating Process-Level Defects and Control Preservation in LLM Coding Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Chaoyi Xue, Chenbo Liu, Dong Sun, Jiawei He, Jie Jia, Xikai Yang, Yapeng Song","submitted_at":"2026-05-18T08:34:48Z","abstract_excerpt":"Existing benchmarks for LLM coding agents mainly evaluate final outcomes, such as task completion, compilation success, and test pass rates. While these metrics are useful for measuring end-task capability, they provide limited visibility into how an execution unfolds and often miss recurrent process-level failures that arise during multi-step operation. We present ProcBench, a benchmark-oriented framework for evaluating coding-agent trajectories through process defects and control preservation. ProcBench organizes execution failures into a reusable ontology, standardizes heterogeneous logs in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20251","kind":"arxiv","version":1},"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/2605.20251/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.20251","created_at":"2026-05-21T00:04:22.765561+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.20251v1","created_at":"2026-05-21T00:04:22.765561+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20251","created_at":"2026-05-21T00:04:22.765561+00:00"},{"alias_kind":"pith_short_12","alias_value":"YMVKTNODOKQE","created_at":"2026-05-21T00:04:22.765561+00:00"},{"alias_kind":"pith_short_16","alias_value":"YMVKTNODOKQES4IW","created_at":"2026-05-21T00:04:22.765561+00:00"},{"alias_kind":"pith_short_8","alias_value":"YMVKTNOD","created_at":"2026-05-21T00:04:22.765561+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6","json":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6.json","graph_json":"https://pith.science/api/pith-number/YMVKTNODOKQES4IWFFMY6UWAO6/graph.json","events_json":"https://pith.science/api/pith-number/YMVKTNODOKQES4IWFFMY6UWAO6/events.json","paper":"https://pith.science/paper/YMVKTNOD"},"agent_actions":{"view_html":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6","download_json":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6.json","view_paper":"https://pith.science/paper/YMVKTNOD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.20251&json=true","fetch_graph":"https://pith.science/api/pith-number/YMVKTNODOKQES4IWFFMY6UWAO6/graph.json","fetch_events":"https://pith.science/api/pith-number/YMVKTNODOKQES4IWFFMY6UWAO6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6/action/storage_attestation","attest_author":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6/action/author_attestation","sign_citation":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6/action/citation_signature","submit_replication":"https://pith.science/pith/YMVKTNODOKQES4IWFFMY6UWAO6/action/replication_record"}},"created_at":"2026-05-21T00:04:22.765561+00:00","updated_at":"2026-05-21T00:04:22.765561+00:00"}