{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CXIIWTYTPUHQFUF53GVLPA7IQN","short_pith_number":"pith:CXIIWTYT","schema_version":"1.0","canonical_sha256":"15d08b4f137d0f02d0bdd9aab783e8836389108fd625c0ee5f4f4851931ad846","source":{"kind":"arxiv","id":"2608.09802","version":1},"attestation_state":"computed","paper":{"title":"SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.CL","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","submitted_at":"2026-08-10T16:23:19Z","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"},"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":"2608.09802","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T16:23:19Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"d7e774c06a7c26b6926ce0f037406d58fc74f5d84c883d4efda13f46bda4627c","abstract_canon_sha256":"62ae4579c2d7f93a8351c2aecbafa461fd5cb95bdb053678ffe985bd7320e4b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T02:25:08.006429Z","signature_b64":"WFJnNtC8pIRCmstdkicZn+SX34gazXOkyjYzwa5worejWN/6rdVYdy2uQUinEMgsv4SCt4gDp1fF6Hdtup3vAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"15d08b4f137d0f02d0bdd9aab783e8836389108fd625c0ee5f4f4851931ad846","last_reissued_at":"2026-08-11T02:25:08.004898Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T02:25:08.004898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.CL","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","submitted_at":"2026-08-10T16:23:19Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.09802","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/2608.09802/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":"2608.09802","created_at":"2026-08-11T02:25:08.005484+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.09802v1","created_at":"2026-08-11T02:25:08.005484+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.09802","created_at":"2026-08-11T02:25:08.005484+00:00"},{"alias_kind":"pith_short_12","alias_value":"CXIIWTYTPUHQ","created_at":"2026-08-11T02:25:08.005484+00:00"},{"alias_kind":"pith_short_16","alias_value":"CXIIWTYTPUHQFUF5","created_at":"2026-08-11T02:25:08.005484+00:00"},{"alias_kind":"pith_short_8","alias_value":"CXIIWTYT","created_at":"2026-08-11T02:25:08.005484+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/CXIIWTYTPUHQFUF53GVLPA7IQN","json":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN.json","graph_json":"https://pith.science/api/pith-number/CXIIWTYTPUHQFUF53GVLPA7IQN/graph.json","events_json":"https://pith.science/api/pith-number/CXIIWTYTPUHQFUF53GVLPA7IQN/events.json","paper":"https://pith.science/paper/CXIIWTYT"},"agent_actions":{"view_html":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN","download_json":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN.json","view_paper":"https://pith.science/paper/CXIIWTYT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.09802&json=true","fetch_graph":"https://pith.science/api/pith-number/CXIIWTYTPUHQFUF53GVLPA7IQN/graph.json","fetch_events":"https://pith.science/api/pith-number/CXIIWTYTPUHQFUF53GVLPA7IQN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN/action/storage_attestation","attest_author":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN/action/author_attestation","sign_citation":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN/action/citation_signature","submit_replication":"https://pith.science/pith/CXIIWTYTPUHQFUF53GVLPA7IQN/action/replication_record"}},"created_at":"2026-08-11T02:25:08.005484+00:00","updated_at":"2026-08-11T02:25:08.005484+00:00"}