{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5MQCWGV5UD27TML62G3J2WBJE3","short_pith_number":"pith:5MQCWGV5","schema_version":"1.0","canonical_sha256":"eb202b1abda0f5f9b17ed1b69d582926d796212fea93c56d899fe4f162b7fc98","source":{"kind":"arxiv","id":"2607.07744","version":1},"attestation_state":"computed","paper":{"title":"PERFOPT-Bench: Evaluating Coding Agents on Software Performance Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Bo Liu, Jiawei Ma, Liangliang Cao, Piaohong Wang, Yingyun Cui, Yi Xie","submitted_at":"2026-07-08T08:32:18Z","abstract_excerpt":"Coding-agent benchmarks have largely measured whether agents can produce functionally correct patches, but production software also demands measurable speedups on real execution targets. Performance optimization is a distinct agentic task: agents must profile executions, diagnose cross-layer bottlenecks, edit code without breaking correctness, and verify that gains are reproducible rather than measurement artifacts. We introduce PERFOPT-Bench, a benchmark for evaluating this full performance-engineering loop. Each task provides a correct but deliberately suboptimal codebase and asks the agent "},"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":"2607.07744","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-08T08:32:18Z","cross_cats_sorted":[],"title_canon_sha256":"ca6bbe20c758dd3a8854c2584e875c708d362d0e53f497c7a121098b284165d7","abstract_canon_sha256":"2f87facfcd23a6ef31745dde50cbc8cbad0c96f2fc3ac933c9ac73a8a3809aec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-10T00:18:47.687640Z","signature_b64":"O8bbtPk9fOvNUgz2t+2Amm5J9TpVFGX+RDOyfhZia1Ja+VmZDEgHE2TtAJVPSZbb9XtdDRudHVH+oxZ5IuZICQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eb202b1abda0f5f9b17ed1b69d582926d796212fea93c56d899fe4f162b7fc98","last_reissued_at":"2026-07-10T00:18:47.687220Z","signature_status":"signed_v1","first_computed_at":"2026-07-10T00:18:47.687220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PERFOPT-Bench: Evaluating Coding Agents on Software Performance Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Bo Liu, Jiawei Ma, Liangliang Cao, Piaohong Wang, Yingyun Cui, Yi Xie","submitted_at":"2026-07-08T08:32:18Z","abstract_excerpt":"Coding-agent benchmarks have largely measured whether agents can produce functionally correct patches, but production software also demands measurable speedups on real execution targets. Performance optimization is a distinct agentic task: agents must profile executions, diagnose cross-layer bottlenecks, edit code without breaking correctness, and verify that gains are reproducible rather than measurement artifacts. We introduce PERFOPT-Bench, a benchmark for evaluating this full performance-engineering loop. Each task provides a correct but deliberately suboptimal codebase and asks the agent "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07744","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/2607.07744/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":"2607.07744","created_at":"2026-07-10T00:18:47.687289+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.07744v1","created_at":"2026-07-10T00:18:47.687289+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07744","created_at":"2026-07-10T00:18:47.687289+00:00"},{"alias_kind":"pith_short_12","alias_value":"5MQCWGV5UD27","created_at":"2026-07-10T00:18:47.687289+00:00"},{"alias_kind":"pith_short_16","alias_value":"5MQCWGV5UD27TML6","created_at":"2026-07-10T00:18:47.687289+00:00"},{"alias_kind":"pith_short_8","alias_value":"5MQCWGV5","created_at":"2026-07-10T00:18:47.687289+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/5MQCWGV5UD27TML62G3J2WBJE3","json":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3.json","graph_json":"https://pith.science/api/pith-number/5MQCWGV5UD27TML62G3J2WBJE3/graph.json","events_json":"https://pith.science/api/pith-number/5MQCWGV5UD27TML62G3J2WBJE3/events.json","paper":"https://pith.science/paper/5MQCWGV5"},"agent_actions":{"view_html":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3","download_json":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3.json","view_paper":"https://pith.science/paper/5MQCWGV5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.07744&json=true","fetch_graph":"https://pith.science/api/pith-number/5MQCWGV5UD27TML62G3J2WBJE3/graph.json","fetch_events":"https://pith.science/api/pith-number/5MQCWGV5UD27TML62G3J2WBJE3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/action/storage_attestation","attest_author":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/action/author_attestation","sign_citation":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/action/citation_signature","submit_replication":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/action/replication_record"}},"created_at":"2026-07-10T00:18:47.687289+00:00","updated_at":"2026-07-10T00:18:47.687289+00:00"}