{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.04023", "kind": "paper", "url": "https://arxiv.org/abs/2606.04023"}, {"kind": "website", "url": "https://anonymous.4open.science/r/CodegenBench-EDE1/"}], "categories": ["Code & Software", "Software & AI Compute", "cs.SE"], "description": {"en": "CodegenBench evaluates LLM-generated parallel code across x86_64, Sunway, and Kunpeng architectures using BLAS routines and specialized kernels. Scoring measures efficiency on each platform."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.04023", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.04023", "title": "CodegenBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.04023", "modality": null, "name": "CodegenBench", "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_codegenbench_b6866f0c", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "f5b5a47d934d3ab269e202e334bd87c502d4946ee10eefec4b9f0e4c42ff4fa9", "relation": "aggregates", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.04023"}, "publisher": null, "released": "2026-06-01", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.04023", "source_url": "https://arxiv.org/abs/2606.04023"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-04023", "source": "claire_radar", "source_benchmark_id": "2606.04023", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Code & Software", "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/2606.04023", "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": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "13e8dba48bbcffc07c2b92d947ad8cb04d4c16124dcef3c28a377910ee231e54", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "CodegenBench evaluates LLM-generated parallel code across x86_64, Sunway, and Kunpeng architectures using BLAS routines and specialized kernels. Scoring measures efficiency on each platform.", "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", "public artifact URL"], "snippet": "To bridge this gap, we introduce CodegenBench, a comprehensive benchmark suite designed to evaluate the generation of efficient parallel code across three distinct hardware platforms: x86_64, Sunway, and Kunpeng."}, "familyId": "bmf_25ab5218d3f7", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_codegenbench_b6866f0c", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.04023", "paper": "https://arxiv.org/abs/2606.04023", "pdf": "https://arxiv.org/pdf/2606.04023", "project": "https://anonymous.4open.science/r/CodegenBench-EDE1/", "report": "https://arxiv.org/abs/2606.04023"}, "metrics": [], "motivation": "While large language models (LLMs) have been extensively evaluated on code generation tasks for general-purpose programming and GPU-accelerated environments (e.g., PyTorch, CUDA), their capabilities in CPU-oriented high-performance computing (HPC) across diverse architectures remain underexplored.", "name": "CodegenBench", "oneLine": "CodegenBench evaluates LLM-generated parallel code across x86_64, Sunway, and Kunpeng architectures using BLAS routines and specialized kernels. Scoring measures efficiency on each platform.", "paperTitle": "CodegenBench: Can LLMs Write Efficient Code Across Architectures?", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.04023", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "aggregates", "releasedAt": "2026-06-01", "source": {"authors": ["Jie Li", "Wenzhao Wu", "Junqi Hu", "Qinrui Zheng", "Bowen Wu", "Juepeng Zheng", "Yutong Lu", "Haohuan Fu"], "categories": ["cs.SE", "cs.AI"], "id": "2606.04023", "title": "CodegenBench: Can LLMs Write Efficient Code Across Architectures?", "type": "arxiv", "url": "https://arxiv.org/abs/2606.04023"}, "sourceUpdatedAt": "2026-06-01T12:55:10Z", "topics": ["cs.SE"], "venueAttempts": [], "whyItMatters": "Addresses the gap in evaluating code generation for CPU-oriented HPC platforms, revealing cross-platform generalization limitations that matter for deploying LLMs in supercomputing contexts."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
