{"slug": "lightweight-multimodal-llm-enabled-cost-effective-defect-grading-of-power", "title": "Lightweight Multimodal LLM-Enabled Cost-Effective Defect Grading of Power Transmission Equipment", "summary": "Researchers have developed a lightweight multimodal large language model framework for defect grading of power transmission equipment, achieving state-of-the-art performance while reducing manual annotation costs. The approach uses in-context learning and chain-of-thought question-answer pairs to fine-tune the Qwen3-VL-8B model, enabling a single lightweight system to handle multiple grading tasks. This advancement addresses class imbalance and expert experience integration challenges in power transmission equipment defect grading, critical for electric energy transmission stability.", "body_md": "arXiv:2605.28822v1 Announce Type: new\nAbstract: Defect grading of power transmission equipment (DGPTE) is crucial to the stability of electric energy transmission. Although existing machine learning methods exhibit strong capabilities in defect detection, they are plagued by difficulties in integrating expert experience and facing class imbalance in more refined defect grading field. To address this issue, this paper introduces a novel defect grading framework based on multimodal large language model (MLLM). Specifically, this approach maximizes the commercial MLLMs' potential of DGPTE through in-context learning and obtains the state-of-te-art (SOTA) model. By sending a secondary request to this model, a small number of chain of thought-based question-answer pairs (Q\\&As) are generated, which effectively reduces the cost of manual annotation. In this way, these high-quality interpretable Q\\&As are used to train Qwen3-VL-8B via Low-Rank Adaption-based supervised fine-tuning (SFT). Experimental results on three DGPTE tasks demonstrate that fine-tuning only the language model layer yields the SOTA performance. Furthermore, multi-task joint fine-tuning verifies the feasibility of handling multiple grading tasks within only a single lightweight MLLM.", "url": "https://wpnews.pro/news/lightweight-multimodal-llm-enabled-cost-effective-defect-grading-of-power", "canonical_source": "https://arxiv.org/abs/2605.28822", "published_at": "2026-05-29 04:00:00+00:00", "updated_at": "2026-05-29 04:23:43.520003+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "computer-vision"], "entities": ["Qwen3-VL-8B", "Low-Rank Adaption", "DGPTE"], "alternates": {"html": "https://wpnews.pro/news/lightweight-multimodal-llm-enabled-cost-effective-defect-grading-of-power", "markdown": "https://wpnews.pro/news/lightweight-multimodal-llm-enabled-cost-effective-defect-grading-of-power.md", "text": "https://wpnews.pro/news/lightweight-multimodal-llm-enabled-cost-effective-defect-grading-of-power.txt", "jsonld": "https://wpnews.pro/news/lightweight-multimodal-llm-enabled-cost-effective-defect-grading-of-power.jsonld"}}