{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:5MQCWGV5UD27TML62G3J2WBJE3","short_pith_number":"pith:5MQCWGV5","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"},"canonical_sha256":"eb202b1abda0f5f9b17ed1b69d582926d796212fea93c56d899fe4f162b7fc98","source":{"kind":"arxiv","id":"2607.07744","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07744","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07744v1","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07744","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"pith_short_12","alias_value":"5MQCWGV5UD27","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"pith_short_16","alias_value":"5MQCWGV5UD27TML6","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"pith_short_8","alias_value":"5MQCWGV5","created_at":"2026-07-10T00:18:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:5MQCWGV5UD27TML62G3J2WBJE3","target":"record","payload":{"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"},"canonical_sha256":"eb202b1abda0f5f9b17ed1b69d582926d796212fea93c56d899fe4f162b7fc98","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"},"source_kind":"arxiv","source_id":"2607.07744","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-10T00:18:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lMi//1brsVkqp+rYSGeLHIJ3hSrBFlwVXksl9KxQl8C+FNS3fb/6BAAG3zu1qchaxyNW+MpYEjYQnAS6fYcWDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T04:26:13.484042Z"},"content_sha256":"c4a6f561925ff4b1cc18a6468ee68baa0b13b78b1842d318fe4629516ca5af2d","schema_version":"1.0","event_id":"sha256:c4a6f561925ff4b1cc18a6468ee68baa0b13b78b1842d318fe4629516ca5af2d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:5MQCWGV5UD27TML62G3J2WBJE3","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-10T00:18:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3z8lBpfUVEWhlcOsXMCZiCGQGSpDRMguyTDIsw1RCA/6tlOxdm2H2fsCYXfrvwhDlOQMYv1V8Shek1bugtaAAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T04:26:13.484555Z"},"content_sha256":"de613f9c5325231b9fdad6fc8f8c7363c8e544d4ac63fed0ca13942a59e53958","schema_version":"1.0","event_id":"sha256:de613f9c5325231b9fdad6fc8f8c7363c8e544d4ac63fed0ca13942a59e53958"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/bundle.json","state_url":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5MQCWGV5UD27TML62G3J2WBJE3/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":"2026-08-20T04:26:13Z","links":{"resolver":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3","bundle":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/bundle.json","state":"https://pith.science/pith/5MQCWGV5UD27TML62G3J2WBJE3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5MQCWGV5UD27TML62G3J2WBJE3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:5MQCWGV5UD27TML62G3J2WBJE3","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":"2f87facfcd23a6ef31745dde50cbc8cbad0c96f2fc3ac933c9ac73a8a3809aec","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-08T08:32:18Z","title_canon_sha256":"ca6bbe20c758dd3a8854c2584e875c708d362d0e53f497c7a121098b284165d7"},"schema_version":"1.0","source":{"id":"2607.07744","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07744","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07744v1","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07744","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"pith_short_12","alias_value":"5MQCWGV5UD27","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"pith_short_16","alias_value":"5MQCWGV5UD27TML6","created_at":"2026-07-10T00:18:47Z"},{"alias_kind":"pith_short_8","alias_value":"5MQCWGV5","created_at":"2026-07-10T00:18:47Z"}],"graph_snapshots":[{"event_id":"sha256:de613f9c5325231b9fdad6fc8f8c7363c8e544d4ac63fed0ca13942a59e53958","target":"graph","created_at":"2026-07-10T00:18:47Z","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/2607.07744/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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 ","authors_text":"Bo Liu, Jiawei Ma, Liangliang Cao, Piaohong Wang, Yingyun Cui, Yi Xie","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-08T08:32:18Z","title":"PERFOPT-Bench: Evaluating Coding Agents on Software Performance Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07744","kind":"arxiv","version":1},"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:c4a6f561925ff4b1cc18a6468ee68baa0b13b78b1842d318fe4629516ca5af2d","target":"record","created_at":"2026-07-10T00:18:47Z","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":"2f87facfcd23a6ef31745dde50cbc8cbad0c96f2fc3ac933c9ac73a8a3809aec","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-08T08:32:18Z","title_canon_sha256":"ca6bbe20c758dd3a8854c2584e875c708d362d0e53f497c7a121098b284165d7"},"schema_version":"1.0","source":{"id":"2607.07744","kind":"arxiv","version":1}},"canonical_sha256":"eb202b1abda0f5f9b17ed1b69d582926d796212fea93c56d899fe4f162b7fc98","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eb202b1abda0f5f9b17ed1b69d582926d796212fea93c56d899fe4f162b7fc98","first_computed_at":"2026-07-10T00:18:47.687220Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-10T00:18:47.687220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"O8bbtPk9fOvNUgz2t+2Amm5J9TpVFGX+RDOyfhZia1Ja+VmZDEgHE2TtAJVPSZbb9XtdDRudHVH+oxZ5IuZICQ==","signature_status":"signed_v1","signed_at":"2026-07-10T00:18:47.687640Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.07744","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c4a6f561925ff4b1cc18a6468ee68baa0b13b78b1842d318fe4629516ca5af2d","sha256:de613f9c5325231b9fdad6fc8f8c7363c8e544d4ac63fed0ca13942a59e53958"],"state_sha256":"28364d78ce5d7e3bf04fdfac472011177143e0da2248b484aa1e73ddd28cbac2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TPy9voRW+1sIss0xLE6iaG+gr48oECzLeKlFIM/tx2H4E3x/M00Kc4rtgwurulA+CXGV6UMp52++yRfsjOKkAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T04:26:13.490196Z","bundle_sha256":"a7f379ff77531b0be3827176b7769a1f290acae562caf1063b044b871ecd7eae"}}