{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WEAVGRBETCRBN5WLR7BU4JFD2W","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":"f2731bff3c16439d8abe83473e28e4c017292233bd476383abf7a463a0eeaa11","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T19:23:51Z","title_canon_sha256":"ea7ea1e305ab50e55f9b31314340bb1b4952f1324b74bb2bd136b9c858766f27"},"schema_version":"1.0","source":{"id":"2506.00172","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.00172","created_at":"2026-07-05T11:13:30Z"},{"alias_kind":"arxiv_version","alias_value":"2506.00172v1","created_at":"2026-07-05T11:13:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.00172","created_at":"2026-07-05T11:13:30Z"},{"alias_kind":"pith_short_12","alias_value":"WEAVGRBETCRB","created_at":"2026-07-05T11:13:30Z"},{"alias_kind":"pith_short_16","alias_value":"WEAVGRBETCRBN5WL","created_at":"2026-07-05T11:13:30Z"},{"alias_kind":"pith_short_8","alias_value":"WEAVGRBE","created_at":"2026-07-05T11:13:30Z"}],"graph_snapshots":[{"event_id":"sha256:f65b53e481af0fd1ddbca25acf5d966c828f56b2f4eb97c351fcb4365298bc32","target":"graph","created_at":"2026-07-05T11:13:30Z","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/2506.00172/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Benchmarks for large language models (LLMs) have predominantly assessed short-horizon, localized reasoning. Existing long-horizon suites (e.g. SWE-bench) rely on manually curated issues, so expanding or tuning difficulty demands expensive human effort and evaluations quickly saturate. However, many real-world tasks, such as software engineering or scientific research, require agents to rapidly comprehend and manipulate novel, complex structures dynamically; evaluating these capabilities requires the ability to construct large and varied sets of problems for agents to solve. We introduce Breakp","authors_text":"Atticus Wang, Jacob Andreas, Kaivalya Hariharan, Uzay Girit","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T19:23:51Z","title":"Breakpoint: Scalable evaluation of system-level reasoning in LLM code agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.00172","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":"sha256:3a9ddf00db1d471a4b5835bc1cf8fe0feb10f1c5b22dd33fff60ae53a665d069","target":"record","created_at":"2026-07-05T11:13:30Z","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":"f2731bff3c16439d8abe83473e28e4c017292233bd476383abf7a463a0eeaa11","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T19:23:51Z","title_canon_sha256":"ea7ea1e305ab50e55f9b31314340bb1b4952f1324b74bb2bd136b9c858766f27"},"schema_version":"1.0","source":{"id":"2506.00172","kind":"arxiv","version":1}},"canonical_sha256":"b10153442498a216f6cb8fc34e24a3d59c6620107fe896dcc517ee900bf83b86","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b10153442498a216f6cb8fc34e24a3d59c6620107fe896dcc517ee900bf83b86","first_computed_at":"2026-07-05T11:13:30.161711Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:30.161711Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NbMA6/HuVviSLm7TlP++F6B2awQyfK4l6TH1PlEv/oeJgrUoEyIYbLocTxKaMrqdB2NvtzxUVhx6wyjdK0YOCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:30.162107Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.00172","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3a9ddf00db1d471a4b5835bc1cf8fe0feb10f1c5b22dd33fff60ae53a665d069","sha256:f65b53e481af0fd1ddbca25acf5d966c828f56b2f4eb97c351fcb4365298bc32"],"state_sha256":"758348756a7f0a16e1ddb56af858b983e4885ecfb88873588052a3d2beb21810"}