{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.11416", "kind": "paper", "url": "https://arxiv.org/abs/2606.11416"}], "authors": ["Yukuan Zhang", "Mengxin Zheng", "Qian Lou"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.11416", "source_url": "https://arxiv.org/abs/2606.11416"}, "categories": ["Language & Knowledge", "Software & AI Compute", "cs.CR"], "description": {"en": "MPC-Patch-Bench evaluates LLM-based code repair on repository-level Secure Multi-Party Computation (MPC) software. It provides 205 verified instances with Fail-to-Pass/Pass-to-Pass tests and a verifier that checks cryptographic safety and numerical fidelity via differential testing and static analysis."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.11416", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.11416", "title": "MPC-Patch-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of Central Florida"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.11416", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.11416", "source_url": "https://openalex.org/W7164359601"}, "key": "claire-radar:2606.11416", "modality": null, "name": "MPC-Patch-Bench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_mpc-patch-bench_574ee054", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "71252f3c18f4c38ebbd428b8f5de1b6f2774c18be0bbda8c23653b9ec22003dc", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.11416"}, "publisher": null, "released": "2026-06-09", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.11416", "source_url": "https://arxiv.org/abs/2606.11416"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-11416", "source": "claire_radar", "source_benchmark_id": "2606.11416", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.11416", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "5f0c20089eb1b773a816e838d7ef44624ec30c955fb68286621f83a0b5ae8754", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "MPC-Patch-Bench evaluates LLM-based code repair on repository-level Secure Multi-Party Computation (MPC) software. It provides 205 verified instances with Fail-to-Pass/Pass-to-Pass tests and a verifier that checks cryptographic safety and numerical fidelity via differential testing and static analysis.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce MPC-Patch-Bench, a repository-level benchmark organised around two frameworks."}, "familyId": "bmf_0c668e75b30f", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_mpc-patch-bench_574ee054", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.11416", "paper": "https://arxiv.org/abs/2606.11416", "pdf": "https://arxiv.org/pdf/2606.11416", "project": null, "report": "https://arxiv.org/abs/2606.11416"}, "metrics": [], "motivation": "Repository-level benchmarks for evaluating Large Language Model (LLM) code repair on Secure Multi-Party Computation (MPC) software do not yet exist, and directly transplanting general-purpose benchmarks such as SWE-bench fails on three structural fronts: (i) MPC repositories are dominated by generic Python infrastructure rather than cryptographic logic; (ii) high-value MPC fixes lack the standardized tests rigid ext…", "name": "MPC-Patch-Bench", "oneLine": "MPC-Patch-Bench evaluates LLM-based code repair on repository-level Secure Multi-Party Computation (MPC) software. It provides 205 verified instances with Fail-to-Pass/Pass-to-Pass tests and a verifier that checks cryptographic safety and numerical fidelity via differential testing and static analysis.", "paperTitle": "MPC-Patch-Bench: Security-Aware LLM Code Patch for Multi-Party Computation", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.11416", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-09", "source": {"authors": ["Yukuan Zhang", "Mengxin Zheng", "Qian Lou"], "categories": ["cs.CR", "cs.AI"], "id": "2606.11416", "title": "MPC-Patch-Bench: Security-Aware LLM Code Patch for Multi-Party Computation", "type": "arxiv", "url": "https://arxiv.org/abs/2606.11416"}, "sourceUpdatedAt": "2026-06-09T20:06:29Z", "topics": ["cs.CR"], "venueAttempts": [], "whyItMatters": "Existing benchmarks lack MPC-aware evaluation for repository-level code repair. MPC-Patch-Bench addresses security and numerical-fidelity gaps, offering a repeatable protocol for assessing LLM agents on real-world MPC tasks, with verification rejected up to 40% of functionally passing patches."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
