{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:7ANK5EOEEQGIOOLGQHHUGVGX23","short_pith_number":"pith:7ANK5EOE","canonical_record":{"source":{"id":"2607.27409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-29T19:28:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c4258ca4a2639142f0f95cf64a8b671aba2f38f1a31beda51ef0d60702935931","abstract_canon_sha256":"69ee5ac363ad6961940441b3af952ea7e4e0a854b127e84c44cf6f2bfafc2dab"},"schema_version":"1.0"},"canonical_sha256":"f81aae91c4240c87396681cf4354d7d6ef0e0bb56ad9cbedd581638d8c07430f","source":{"kind":"arxiv","id":"2607.27409","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.27409","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"arxiv_version","alias_value":"2607.27409v1","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27409","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"pith_short_12","alias_value":"7ANK5EOEEQGI","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"pith_short_16","alias_value":"7ANK5EOEEQGIOOLG","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"pith_short_8","alias_value":"7ANK5EOE","created_at":"2026-07-31T00:11:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:7ANK5EOEEQGIOOLGQHHUGVGX23","target":"record","payload":{"canonical_record":{"source":{"id":"2607.27409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-29T19:28:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c4258ca4a2639142f0f95cf64a8b671aba2f38f1a31beda51ef0d60702935931","abstract_canon_sha256":"69ee5ac363ad6961940441b3af952ea7e4e0a854b127e84c44cf6f2bfafc2dab"},"schema_version":"1.0"},"canonical_sha256":"f81aae91c4240c87396681cf4354d7d6ef0e0bb56ad9cbedd581638d8c07430f","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f81aae91c4240c87396681cf4354d7d6ef0e0bb56ad9cbedd581638d8c07430f","last_reissued_at":"2026-07-31T00:11:44.140129Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-31T00:11:44.140129Z"},"source_kind":"arxiv","source_id":"2607.27409","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":"2026-07-31T00:11:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TBvcpYjSZpcoU4f4riioFuPbwmClRS1sGOehk1KJxLZrNLALAU0xvAWrrK0aSaTcBqWixsOd0JpXw8uhnqCCCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T05:37:13.552046Z"},"content_sha256":"49865954199b57666f0789dd79281fb60ee447c67139a55bd29d9746d0238ec0","schema_version":"1.0","event_id":"sha256:49865954199b57666f0789dd79281fb60ee447c67139a55bd29d9746d0238ec0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:7ANK5EOEEQGIOOLGQHHUGVGX23","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SWE-NFI: Studying and Benchmarking Coding Agents for Non-Functional Improvements","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Boyuan Chen, Haonan Zhang, He Yang Yuan, Junkai Chen, Pengyu Xue, Weiyi Shang, Xin Wang, Zhenhao Li, Zishuo Ding","submitted_at":"2026-07-29T19:28:32Z","abstract_excerpt":"Although coding agents have achieved impressive performance on correctness-oriented benchmarks, their ability to make behavior-preserving non-functional improvements (NFIs) remains underexplored. In real-world software development, developers continuously improve software quality without changing observable behavior, yet existing benchmarks primarily evaluate functional correctness and provide limited support for assessing these non-functional improvements.\n  In this paper, we present SWE-NFI, a benchmark for evaluating coding agents on NFIs beyond functional correctness. Our benchmark contain"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27409","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/2607.27409/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":"2026-07-31T00:11:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0BqPcitXj/YgUuv3/JYDfiqf5jznZMoQXOjcslrFiFgah0STq//TG6CYuA5FN8TQPeWEcT1ssjaHb8q4+RE/BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T05:37:13.552423Z"},"content_sha256":"c839a603813eab7a2607e196db71060a55bc991ecb889d18a2132bb4df1d938a","schema_version":"1.0","event_id":"sha256:c839a603813eab7a2607e196db71060a55bc991ecb889d18a2132bb4df1d938a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7ANK5EOEEQGIOOLGQHHUGVGX23/bundle.json","state_url":"https://pith.science/pith/7ANK5EOEEQGIOOLGQHHUGVGX23/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7ANK5EOEEQGIOOLGQHHUGVGX23/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":"2026-08-04T05:37:13Z","links":{"resolver":"https://pith.science/pith/7ANK5EOEEQGIOOLGQHHUGVGX23","bundle":"https://pith.science/pith/7ANK5EOEEQGIOOLGQHHUGVGX23/bundle.json","state":"https://pith.science/pith/7ANK5EOEEQGIOOLGQHHUGVGX23/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7ANK5EOEEQGIOOLGQHHUGVGX23/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7ANK5EOEEQGIOOLGQHHUGVGX23","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":"69ee5ac363ad6961940441b3af952ea7e4e0a854b127e84c44cf6f2bfafc2dab","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-29T19:28:32Z","title_canon_sha256":"c4258ca4a2639142f0f95cf64a8b671aba2f38f1a31beda51ef0d60702935931"},"schema_version":"1.0","source":{"id":"2607.27409","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.27409","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"arxiv_version","alias_value":"2607.27409v1","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27409","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"pith_short_12","alias_value":"7ANK5EOEEQGI","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"pith_short_16","alias_value":"7ANK5EOEEQGIOOLG","created_at":"2026-07-31T00:11:44Z"},{"alias_kind":"pith_short_8","alias_value":"7ANK5EOE","created_at":"2026-07-31T00:11:44Z"}],"graph_snapshots":[{"event_id":"sha256:c839a603813eab7a2607e196db71060a55bc991ecb889d18a2132bb4df1d938a","target":"graph","created_at":"2026-07-31T00:11:44Z","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/2607.27409/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Although coding agents have achieved impressive performance on correctness-oriented benchmarks, their ability to make behavior-preserving non-functional improvements (NFIs) remains underexplored. In real-world software development, developers continuously improve software quality without changing observable behavior, yet existing benchmarks primarily evaluate functional correctness and provide limited support for assessing these non-functional improvements.\n  In this paper, we present SWE-NFI, a benchmark for evaluating coding agents on NFIs beyond functional correctness. Our benchmark contain","authors_text":"Boyuan Chen, Haonan Zhang, He Yang Yuan, Junkai Chen, Pengyu Xue, Weiyi Shang, Xin Wang, Zhenhao Li, Zishuo Ding","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-29T19:28:32Z","title":"SWE-NFI: Studying and Benchmarking Coding Agents for Non-Functional Improvements"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27409","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":"sha256:49865954199b57666f0789dd79281fb60ee447c67139a55bd29d9746d0238ec0","target":"record","created_at":"2026-07-31T00:11:44Z","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":"69ee5ac363ad6961940441b3af952ea7e4e0a854b127e84c44cf6f2bfafc2dab","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-29T19:28:32Z","title_canon_sha256":"c4258ca4a2639142f0f95cf64a8b671aba2f38f1a31beda51ef0d60702935931"},"schema_version":"1.0","source":{"id":"2607.27409","kind":"arxiv","version":1}},"canonical_sha256":"f81aae91c4240c87396681cf4354d7d6ef0e0bb56ad9cbedd581638d8c07430f","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f81aae91c4240c87396681cf4354d7d6ef0e0bb56ad9cbedd581638d8c07430f","first_computed_at":"2026-07-31T00:11:44.140129Z","kind":"pith_receipt","last_reissued_at":"2026-07-31T00:11:44.140129Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2607.27409","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:49865954199b57666f0789dd79281fb60ee447c67139a55bd29d9746d0238ec0","sha256:c839a603813eab7a2607e196db71060a55bc991ecb889d18a2132bb4df1d938a"],"state_sha256":"d49a2c6531454d0e1f16e909e3a0bfaa1b36d823f81160041d06801198374133"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3QspnJAeSsSy61ZlOIMyEfSdxpDIp05Q978lcVgGE3bVCiA5gtBK/kmDvUiP4rzoYaHzjXYla9aqIz4aVGmpCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T05:37:13.554935Z","bundle_sha256":"6c357b26157775a794788408709ed4623e24a54b303af923816b9d4ee69a36eb"}}