{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:CXIIWTYTPUHQFUF53GVLPA7IQN","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":"62ae4579c2d7f93a8351c2aecbafa461fd5cb95bdb053678ffe985bd7320e4b1","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T16:23:19Z","title_canon_sha256":"d7e774c06a7c26b6926ce0f037406d58fc74f5d84c883d4efda13f46bda4627c"},"schema_version":"1.0","source":{"id":"2608.09802","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.09802","created_at":"2026-08-11T02:25:08Z"},{"alias_kind":"arxiv_version","alias_value":"2608.09802v1","created_at":"2026-08-11T02:25:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.09802","created_at":"2026-08-11T02:25:08Z"},{"alias_kind":"pith_short_12","alias_value":"CXIIWTYTPUHQ","created_at":"2026-08-11T02:25:08Z"},{"alias_kind":"pith_short_16","alias_value":"CXIIWTYTPUHQFUF5","created_at":"2026-08-11T02:25:08Z"},{"alias_kind":"pith_short_8","alias_value":"CXIIWTYT","created_at":"2026-08-11T02:25:08Z"}],"graph_snapshots":[{"event_id":"sha256:b83a37f89541cdb4b1aa60c4216b8ad4d9d196c7798de4b3d6c10783a6e5a6e4","target":"graph","created_at":"2026-08-11T02:25:08Z","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/2608.09802/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As AI coding agents take on increasingly complex, long-horizon software engineering tasks, existing benchmarks are rapidly saturating and their evaluation quality has come under serious scrutiny: a recent audit found that nearly 60% of unsolved SWE-bench Verified instances contain flawed tests -- either overly narrow tests that reject correct solutions or overly broad tests that check unstated requirements -- and that frontier models can verbatim reproduce gold patches from training data. Code refactoring, which requires coordinated, behavior-preserving changes across many files, offers a subs","authors_text":"Jialun Cao, Jinghan Xu, Kai Cai, Kelin Fu, Lei Zhang, Shilin He, Shing-Chi Cheung, Siyu Ye, Terry Yue Zhuo, Tianyu Liu, Wenhao Zeng, Xiaodong Gu, Yue Liu, Yuling Shi, Zelin Zhao","cross_cats":["cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T16:23:19Z","title":"SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.09802","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":"sha256:447b36c9252f7f9ab4a96acaa25c40a09bbbeb5ff6ab5647ba7e27c2c0a49d87","target":"record","created_at":"2026-08-11T02:25:08Z","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":"62ae4579c2d7f93a8351c2aecbafa461fd5cb95bdb053678ffe985bd7320e4b1","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T16:23:19Z","title_canon_sha256":"d7e774c06a7c26b6926ce0f037406d58fc74f5d84c883d4efda13f46bda4627c"},"schema_version":"1.0","source":{"id":"2608.09802","kind":"arxiv","version":1}},"canonical_sha256":"15d08b4f137d0f02d0bdd9aab783e8836389108fd625c0ee5f4f4851931ad846","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"15d08b4f137d0f02d0bdd9aab783e8836389108fd625c0ee5f4f4851931ad846","first_computed_at":"2026-08-11T02:25:08.004898Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-11T02:25:08.004898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WFJnNtC8pIRCmstdkicZn+SX34gazXOkyjYzwa5worejWN/6rdVYdy2uQUinEMgsv4SCt4gDp1fF6Hdtup3vAw==","signature_status":"signed_v1","signed_at":"2026-08-11T02:25:08.006429Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.09802","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:447b36c9252f7f9ab4a96acaa25c40a09bbbeb5ff6ab5647ba7e27c2c0a49d87","sha256:b83a37f89541cdb4b1aa60c4216b8ad4d9d196c7798de4b3d6c10783a6e5a6e4"],"state_sha256":"cafa70f9276cd578a9adb6a2e3cb7fa5d0e7a4c444ae10e313861b2445ad52e9"}