{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LEK5HMXNWMMINOIIJVWWSTOMRP","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":"0c10cdb179d49f831043fca9406faedd87723d20923a00511dfb07d548b07d17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SE","submitted_at":"2024-11-27T03:25:44Z","title_canon_sha256":"590b21b77ec4d974350561b22a6d698e9997e2d14e685e601814ab7d73fed038"},"schema_version":"1.0","source":{"id":"2411.18019","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.18019","created_at":"2026-07-05T09:41:10Z"},{"alias_kind":"arxiv_version","alias_value":"2411.18019v1","created_at":"2026-07-05T09:41:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.18019","created_at":"2026-07-05T09:41:10Z"},{"alias_kind":"pith_short_12","alias_value":"LEK5HMXNWMMI","created_at":"2026-07-05T09:41:10Z"},{"alias_kind":"pith_short_16","alias_value":"LEK5HMXNWMMINOII","created_at":"2026-07-05T09:41:10Z"},{"alias_kind":"pith_short_8","alias_value":"LEK5HMXN","created_at":"2026-07-05T09:41:10Z"}],"graph_snapshots":[{"event_id":"sha256:0373fd5141299218156a8259048cb7e03b44e377cef8f1266bbfed5c98636d47","target":"graph","created_at":"2026-07-05T09:41:10Z","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/2411.18019/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Automatically resolving software issues is crucial for software development in practice, impacting the software quality and user experience. The process of resolving real-world issues encompasses tasks such as question-answering (QA), fault localization, and code editing. Existing benchmarks such as HumanEval fall short in their ability to assess LLMs' proficiency in solving issues within a codebase. Although benchmarks like SWE-Bench are designed to evaluate the LLMs' capability to handle real-world GitHub issues, the end-to-end evaluation method cannot provide granular insights on the perfor","authors_text":"Bo Jiang, Chao Peng, Cuiyun Gao, Jingyi Ren, Pengfei Gao, Qinyun Wu, Ruida Hu, Xiangxin Meng, Xinchen Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SE","submitted_at":"2024-11-27T03:25:44Z","title":"A Real-World Benchmark for Evaluating Fine-Grained Issue Solving Capabilities of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.18019","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":"sha256:7198722e686318da8a6663256f0565c2a060d8b8b0c60efeb6047abe96eb594b","target":"record","created_at":"2026-07-05T09:41:10Z","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":"0c10cdb179d49f831043fca9406faedd87723d20923a00511dfb07d548b07d17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SE","submitted_at":"2024-11-27T03:25:44Z","title_canon_sha256":"590b21b77ec4d974350561b22a6d698e9997e2d14e685e601814ab7d73fed038"},"schema_version":"1.0","source":{"id":"2411.18019","kind":"arxiv","version":1}},"canonical_sha256":"5915d3b2edb31886b9084d6d694dcc8bd040cc2407d403dce31871c32e83660e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5915d3b2edb31886b9084d6d694dcc8bd040cc2407d403dce31871c32e83660e","first_computed_at":"2026-07-05T09:41:10.488332Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:41:10.488332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6lBVqPlb2nF5tQVkniup5h4B7bABr2G7wkf7MUGP9SIbJ0N6oMqz69hD4dhbuGlgZ3shBjYJ2yDXn9gGeK73Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:41:10.488816Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.18019","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7198722e686318da8a6663256f0565c2a060d8b8b0c60efeb6047abe96eb594b","sha256:0373fd5141299218156a8259048cb7e03b44e377cef8f1266bbfed5c98636d47"],"state_sha256":"af4413a229e375b67309856851e37072398c5db841d6bb6a516a171721f21a47"}