{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZSKGLUENOJXIRCATTTCOV2CGYT","short_pith_number":"pith:ZSKGLUEN","canonical_record":{"source":{"id":"2503.07832","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-10T20:23:24Z","cross_cats_sorted":["cs.CL","cs.LG","cs.SE"],"title_canon_sha256":"dbab28fb36c2a2c17e4a9e946c70d283906c1bacd03d025557e55c1cefc8de56","abstract_canon_sha256":"d45cd72c75b596bd43ff2b4c2a7c2136d616647a2d645ffe86875a21d196b799"},"schema_version":"1.0"},"canonical_sha256":"cc9465d08d726e8888139cc4eae846c4ff54199a35422793f54b9255bad3c1b6","source":{"kind":"arxiv","id":"2503.07832","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.07832","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"arxiv_version","alias_value":"2503.07832v1","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.07832","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"pith_short_12","alias_value":"ZSKGLUENOJXI","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"pith_short_16","alias_value":"ZSKGLUENOJXIRCAT","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"pith_short_8","alias_value":"ZSKGLUEN","created_at":"2026-07-05T10:28:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZSKGLUENOJXIRCATTTCOV2CGYT","target":"record","payload":{"canonical_record":{"source":{"id":"2503.07832","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-10T20:23:24Z","cross_cats_sorted":["cs.CL","cs.LG","cs.SE"],"title_canon_sha256":"dbab28fb36c2a2c17e4a9e946c70d283906c1bacd03d025557e55c1cefc8de56","abstract_canon_sha256":"d45cd72c75b596bd43ff2b4c2a7c2136d616647a2d645ffe86875a21d196b799"},"schema_version":"1.0"},"canonical_sha256":"cc9465d08d726e8888139cc4eae846c4ff54199a35422793f54b9255bad3c1b6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:28:44.801469Z","signature_b64":"gEnLJwSK6p5UqYnbBEiZORsj8pgCB4KXoHm4Ps104ZHXiRRuk6HBYoj7do/INHhF0vrqpd1HfMfYQFFEhggsDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cc9465d08d726e8888139cc4eae846c4ff54199a35422793f54b9255bad3c1b6","last_reissued_at":"2026-07-05T10:28:44.800968Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:28:44.800968Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.07832","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":"2026-07-05T10:28:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yf4o/7fL3swPQj2RpFinv45a2RGOtI4ExaqauLAAs2pULo3aN8MU6CwHSHX4Dm2Io+YTZrMfYo225rQdR6ZyAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T16:21:51.700885Z"},"content_sha256":"919eefda7625fbdfbed019a540c265e2e312d4a5283d8072861ce76e91e64a2f","schema_version":"1.0","event_id":"sha256:919eefda7625fbdfbed019a540c265e2e312d4a5283d8072861ce76e91e64a2f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZSKGLUENOJXIRCATTTCOV2CGYT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RefactorBench: Evaluating Stateful Reasoning in Language Agents Through Code","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.SE"],"primary_cat":"cs.AI","authors_text":"Dhruv Gautam, Jinu Jang, Neel Sundaresan, Roshanak Zilouchian Moghaddam, Spandan Garg","submitted_at":"2025-03-10T20:23:24Z","abstract_excerpt":"Recent advances in language model (LM) agents and function calling have enabled autonomous, feedback-driven systems to solve problems across various digital domains. To better understand the unique limitations of LM agents, we introduce RefactorBench, a benchmark consisting of 100 large handcrafted multi-file refactoring tasks in popular open-source repositories. Solving tasks within RefactorBench requires thorough exploration of dependencies across multiple files and strong adherence to relevant instructions. Every task is defined by 3 natural language instructions of varying specificity and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.07832","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/2503.07832/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":"2026-07-05T10:28:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ziKSRHGjOcdsL3RctPrqn0Hr5NKqO4FsByVICqVq2S9RILWQKQ7DbgFqUgfXawnz4xNWAn11T22i3XxC+TGNBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T16:21:51.701266Z"},"content_sha256":"10adad395d2d434595becde98c44512feb2130d2c1d8f910ea76ffc97a3bce51","schema_version":"1.0","event_id":"sha256:10adad395d2d434595becde98c44512feb2130d2c1d8f910ea76ffc97a3bce51"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZSKGLUENOJXIRCATTTCOV2CGYT/bundle.json","state_url":"https://pith.science/pith/ZSKGLUENOJXIRCATTTCOV2CGYT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZSKGLUENOJXIRCATTTCOV2CGYT/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":"2026-07-22T16:21:51Z","links":{"resolver":"https://pith.science/pith/ZSKGLUENOJXIRCATTTCOV2CGYT","bundle":"https://pith.science/pith/ZSKGLUENOJXIRCATTTCOV2CGYT/bundle.json","state":"https://pith.science/pith/ZSKGLUENOJXIRCATTTCOV2CGYT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZSKGLUENOJXIRCATTTCOV2CGYT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZSKGLUENOJXIRCATTTCOV2CGYT","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":"d45cd72c75b596bd43ff2b4c2a7c2136d616647a2d645ffe86875a21d196b799","cross_cats_sorted":["cs.CL","cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-10T20:23:24Z","title_canon_sha256":"dbab28fb36c2a2c17e4a9e946c70d283906c1bacd03d025557e55c1cefc8de56"},"schema_version":"1.0","source":{"id":"2503.07832","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.07832","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"arxiv_version","alias_value":"2503.07832v1","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.07832","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"pith_short_12","alias_value":"ZSKGLUENOJXI","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"pith_short_16","alias_value":"ZSKGLUENOJXIRCAT","created_at":"2026-07-05T10:28:44Z"},{"alias_kind":"pith_short_8","alias_value":"ZSKGLUEN","created_at":"2026-07-05T10:28:44Z"}],"graph_snapshots":[{"event_id":"sha256:10adad395d2d434595becde98c44512feb2130d2c1d8f910ea76ffc97a3bce51","target":"graph","created_at":"2026-07-05T10:28:44Z","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/2503.07832/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in language model (LM) agents and function calling have enabled autonomous, feedback-driven systems to solve problems across various digital domains. To better understand the unique limitations of LM agents, we introduce RefactorBench, a benchmark consisting of 100 large handcrafted multi-file refactoring tasks in popular open-source repositories. Solving tasks within RefactorBench requires thorough exploration of dependencies across multiple files and strong adherence to relevant instructions. Every task is defined by 3 natural language instructions of varying specificity and ","authors_text":"Dhruv Gautam, Jinu Jang, Neel Sundaresan, Roshanak Zilouchian Moghaddam, Spandan Garg","cross_cats":["cs.CL","cs.LG","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-10T20:23:24Z","title":"RefactorBench: Evaluating Stateful Reasoning in Language Agents Through Code"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.07832","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":"sha256:919eefda7625fbdfbed019a540c265e2e312d4a5283d8072861ce76e91e64a2f","target":"record","created_at":"2026-07-05T10:28:44Z","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":"d45cd72c75b596bd43ff2b4c2a7c2136d616647a2d645ffe86875a21d196b799","cross_cats_sorted":["cs.CL","cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-10T20:23:24Z","title_canon_sha256":"dbab28fb36c2a2c17e4a9e946c70d283906c1bacd03d025557e55c1cefc8de56"},"schema_version":"1.0","source":{"id":"2503.07832","kind":"arxiv","version":1}},"canonical_sha256":"cc9465d08d726e8888139cc4eae846c4ff54199a35422793f54b9255bad3c1b6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cc9465d08d726e8888139cc4eae846c4ff54199a35422793f54b9255bad3c1b6","first_computed_at":"2026-07-05T10:28:44.800968Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:28:44.800968Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gEnLJwSK6p5UqYnbBEiZORsj8pgCB4KXoHm4Ps104ZHXiRRuk6HBYoj7do/INHhF0vrqpd1HfMfYQFFEhggsDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:28:44.801469Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.07832","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:919eefda7625fbdfbed019a540c265e2e312d4a5283d8072861ce76e91e64a2f","sha256:10adad395d2d434595becde98c44512feb2130d2c1d8f910ea76ffc97a3bce51"],"state_sha256":"ae17f52a5d92bebf15cff7b1996e189b50d6c1170d81aeca7f0ee68fb514691f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ePwTkVKIZBJGYRvtdvVa74zvmIaCj1X8p4lPdhgOZYv+jBilosWhP9IT5OnuOQJrjDuT5eOE52bullFJIprHBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-22T16:21:51.703570Z","bundle_sha256":"3645e83a1a0661dd1eed52fd41734f68372811e5f63f9ac0aff72bd1dfa69b49"}}