{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.31478", "kind": "paper", "url": "https://arxiv.org/abs/2605.31478"}], "authors": ["Hui Wu", "Xiaoyang Wang", "Zhong Fan"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.31478", "source_url": "https://arxiv.org/abs/2605.31478"}, "categories": ["Code", "Code & Software", "Code generation", "Factuality", "Software & AI Compute"], "description": {"en": "PowerCodeBench is a benchmark generator for power system code generation, paired with an intervention method, but no artifacts are provided in this article."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.31478", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.31478", "title": "PowerCodeBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2605.31478", "modality": null, "name": "PowerCodeBench", "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_powercodebench_c0ce0f67", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "32432a33e35f24bdcc4bd7c65aae4e0fde46ab597670e1c4b3d9823deb03958b", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.31478"}, "publisher": null, "released": "2026-05-29", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.31478", "source_url": "https://arxiv.org/abs/2605.31478"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-31478", "source": "claire_radar", "source_benchmark_id": "2605.31478", "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/2605.31478", "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": ["Code generation", "Factuality"], "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": "db350dac8e1dfdd93cb56df1e7b068a6448279e388e3944139dc9fe5222260fc", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "PowerCodeBench is a benchmark generator for power system code generation, paired with an intervention method, but no artifacts are provided in this article.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce PowerCodeBench, an execution-validated benchmark generator that pairs natural-language operator queries with pandapower code and numerical ground truth; an L0-L3 documentation-driven probing procedure that measures per-model API knowledge profiles; and a boundary-aware intervention that combines query-side API demand estimation with targeted proactive documentation injection and routed reactive correction."}, "familyId": "bmf_dc78daa563c7", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_powercodebench_c0ce0f67", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2605.31478", "paper": "https://arxiv.org/abs/2605.31478", "pdf": "https://arxiv.org/pdf/2605.31478", "project": null, "report": "https://arxiv.org/abs/2605.31478"}, "metrics": [], "motivation": "Large language models (LLMs) are increasingly used to automate power-system analysis, but many utilities and energy-research labs require on-premise serving for confidentiality, regulatory, reproducibility, and cost reasons.", "name": "PowerCodeBench", "oneLine": "PowerCodeBench is a benchmark generator for power system code generation, paired with an intervention method, but no artifacts are provided in this article.", "paperTitle": "Knowledge Boundary Probing and Demand-Guided Intervention for LLM-Based Power System Code Generation", "primaryDomain": "Software & AI Compute", "publication": {"evidence": "Advanced Engineering Informatics 77 (2027) 105328", "evidenceLevel": "strong-author-metadata", "evidenceUrl": "https://arxiv.org/abs/2605.31478", "source": "arxiv-journal-reference", "status": "publication_reported", "venue": "Advanced Engineering Informatics 77 (2027) 105328", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [{"evidence": [{"level": "strong-author-metadata", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Advanced Engineering Informatics 77 (2027) 105328", "sourceType": "arxiv-journal-reference", "sourceUrl": "https://arxiv.org/abs/2605.31478"}], "publicationStatus": "published", "venueName": "Advanced Engineering Informatics 77 (2027) 105328"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-05-29", "source": {"authors": ["Hui Wu", "Xiaoyang Wang", "Zhong Fan"], "categories": ["cs.SE", "cs.CL", "cs.SY", "eess.SY"], "id": "2605.31478", "title": "Knowledge Boundary Probing and Demand-Guided Intervention for LLM-Based Power System Code Generation", "type": "arxiv", "url": "https://arxiv.org/abs/2605.31478"}, "sourceUpdatedAt": "2026-05-29T16:06:34Z", "topics": ["Code"], "venueAttempts": [], "whyItMatters": "It addresses reliability of open-weight models for on-premise deployment, but the lack of a release prevents external use."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
