Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wu, Hui, Wang, Xiaoyang, Fan, Zhong
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.31478
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917548175392768
author Wu, Hui
Wang, Xiaoyang
Fan, Zhong
author_facet Wu, Hui
Wang, Xiaoyang
Fan, Zhong
contents 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. This makes the reliability of open-weight models a deployment issue. We show that first-pass failures in power-system code generation are dominated not by reasoning alone, but by structured API-knowledge boundary errors: hallucinated function names, misused parameters, and mishandled result tables in versioned simulation libraries. 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. On a 2,000-task frozen release, we evaluate ten open-weight LLMs (1.5B-480B parameters) and four commercial mid-tier APIs. The intervention improves every evaluated open-weight model of at least 7B parameters and every commercial API by 32 to 56 accuracy points. Open-weight models in the 70B-120B range match the commercial mid-tier accuracy range, while Llama-3.1-405B and Qwen3-Coder-480B lead the panel. The targeted prompts preserve the full-context accuracy ceiling while using 41% of the prompt-token cost. The result is an accuracy-side, deployment-time path toward reliable on-premise LLM assistance for grid-analysis workflows without fine-tuning or cloud inference.
format Preprint
id arxiv_https___arxiv_org_abs_2605_31478
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Knowledge Boundary Probing and Demand-Guided Intervention for LLM-Based Power System Code Generation
Wu, Hui
Wang, Xiaoyang
Fan, Zhong
Software Engineering
Computation and Language
Systems and Control
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. This makes the reliability of open-weight models a deployment issue. We show that first-pass failures in power-system code generation are dominated not by reasoning alone, but by structured API-knowledge boundary errors: hallucinated function names, misused parameters, and mishandled result tables in versioned simulation libraries. 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. On a 2,000-task frozen release, we evaluate ten open-weight LLMs (1.5B-480B parameters) and four commercial mid-tier APIs. The intervention improves every evaluated open-weight model of at least 7B parameters and every commercial API by 32 to 56 accuracy points. Open-weight models in the 70B-120B range match the commercial mid-tier accuracy range, while Llama-3.1-405B and Qwen3-Coder-480B lead the panel. The targeted prompts preserve the full-context accuracy ceiling while using 41% of the prompt-token cost. The result is an accuracy-side, deployment-time path toward reliable on-premise LLM assistance for grid-analysis workflows without fine-tuning or cloud inference.
title Knowledge Boundary Probing and Demand-Guided Intervention for LLM-Based Power System Code Generation
topic Software Engineering
Computation and Language
Systems and Control
url https://arxiv.org/abs/2605.31478