{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:3ERMXYEUNL7GILRH3LOW5IB4EF","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":"776be9b3b336d60fb629c8e38997a3ccbad529873805d864e7bafc74cce5657b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MA","submitted_at":"2026-05-28T17:54:01Z","title_canon_sha256":"9c4a98552721c5dcdddca6b5c5e4965b86619c84403e8417b1dc013a5ab5bd08"},"schema_version":"1.0","source":{"id":"2605.30314","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.30314","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"arxiv_version","alias_value":"2605.30314v1","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30314","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"pith_short_12","alias_value":"3ERMXYEUNL7G","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"pith_short_16","alias_value":"3ERMXYEUNL7GILRH","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"pith_short_8","alias_value":"3ERMXYEU","created_at":"2026-05-29T02:06:16Z"}],"graph_snapshots":[{"event_id":"sha256:a0fcfea48acd21c07dee4f0351c5f6ec05506627119c4aa03032df1f55c7e8fa","target":"graph","created_at":"2026-05-29T02:06:16Z","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/2605.30314/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Software engineering (SWE) agents are transitioning from code generation to full software development lifecycle automation. A critical phase in this lifecycle is specification design: transforming initial proposals into carefully considered requirements through expert review. Existing benchmarks such as SWE-Bench are implementation-focused by measuring the agent's ability to generate code given fixed, precise design requirements. This formulation assumes specifications are correct and complete. In real-world complex and critical software systems, initial specifications are often incomplete and","authors_text":"Anand Jayarajan, Gennady Pekhimenko, Grant Hamblin, Kevin Song, Nandita Vijaykumar, Sihang Liu, Zhanda Zhu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MA","submitted_at":"2026-05-28T17:54:01Z","title":"SpecBench: Evaluating Specification-Level Reasoning for Software Engineering LLM Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30314","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":"sha256:889f802cc5b8a05bca595fe5c31fbe36a1156aa807d8e47fa32d62c1676c092e","target":"record","created_at":"2026-05-29T02:06:16Z","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":"776be9b3b336d60fb629c8e38997a3ccbad529873805d864e7bafc74cce5657b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MA","submitted_at":"2026-05-28T17:54:01Z","title_canon_sha256":"9c4a98552721c5dcdddca6b5c5e4965b86619c84403e8417b1dc013a5ab5bd08"},"schema_version":"1.0","source":{"id":"2605.30314","kind":"arxiv","version":1}},"canonical_sha256":"d922cbe0946afe642e27dadd6ea03c2150661965adb300971c748d37d3e3e814","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d922cbe0946afe642e27dadd6ea03c2150661965adb300971c748d37d3e3e814","first_computed_at":"2026-05-29T02:06:16.233553Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T02:06:16.233553Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UTkl8o9z4mDT7OO5mKg+73+0ud5xGTaWRegizz8GV+DUCfSCLotBO5I44mdgolr78NC6m3KqstYYK4z2U0XdDg==","signature_status":"signed_v1","signed_at":"2026-05-29T02:06:16.233942Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.30314","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:889f802cc5b8a05bca595fe5c31fbe36a1156aa807d8e47fa32d62c1676c092e","sha256:a0fcfea48acd21c07dee4f0351c5f6ec05506627119c4aa03032df1f55c7e8fa"],"state_sha256":"306c9f224cf283144c19e51e23f98f621880ceaa2b3460dd31f58446d4f33720"}