# A Hybrid Vision-Language Architecture for Automated Defect Reasoning and Report Generation in Industrial Inspection

arXiv: https://arxiv.org/abs/2605.26533 · HTML: https://mlsanity.com/paper/2605.26533/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning.
- Builds on 13 of its 45 references that are in the map.
- First posted on arXiv 26 May 2026.

## Placed in

- LLM reasoning / [Chain-of-thought reasoning for driving](https://mlsanity.com/lr/chain-of-thought-reasoning-for-driving/)

## Builds on (most cited first, 13 of 13)

- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models](https://mlsanity.com/paper/2308.15366/) (2023)
- [A Review on Edge Large Language Models: Design, Execution, and Applications](https://mlsanity.com/paper/2410.11845/) (2024)
- [YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection](https://mlsanity.com/paper/2509.25164/) (2025)
- [Effectiveness Assessment of Recent Large Vision-Language Models](https://mlsanity.com/paper/2403.04306/) (2024)
- [FabGPT: An Efficient Large Multimodal Model for Complex Wafer Defect Knowledge Queries](https://mlsanity.com/paper/2407.10810/) (2024)
- [Image Synthesis under Limited Data: A Survey and Taxonomy](https://mlsanity.com/paper/2307.16879/) (2023)
- [Identification of Surface Defects on Solar PV Panels and Wind Turbine Blades using Attention based Deep Learning Model](https://mlsanity.com/paper/2211.15374/) (2022)
- [MAU-GPT: Enhancing Multi-type Industrial Anomaly Understanding via Anomaly-aware and Generalist Experts Adaptation](https://mlsanity.com/paper/2602.07011/) (2026)
- [LLM-Guided Evolution: An Autonomous Model Optimization for Object Detection](https://mlsanity.com/paper/2504.02280/) (2025)
- [A domain adaptation neural network for digital twin-supported fault diagnosis](https://mlsanity.com/paper/2505.21046/) (2025)
- [Cognitive-YOLO: LLM-Driven Architecture Synthesis from First Principles of Data for Object Detection](https://mlsanity.com/paper/2512.12281/) (2025)

## Abstract (arXiv)

Automated industrial inspection requires both precise defect localization and structured maintenance report generation; in current practice these tasks are handled separately, with linguistic interpretation left to human experts. This paper describes a decoupled, edge-deployable pipeline for wind turbine blade inspection built from three components that each handle a distinct sub-task. The Eyes a YOLO26-x-obb oriented bounding-box detector localizes defects at dataset-native resolution. The Bridge a deterministic, parameter-free encoding module maps each detected bounding box to grid-referenced spatial tokens embedded in a structured prompt. The Brain a 4-bit quantized Qwen-2.5-1.5B model adapted with Quantized Low-Rank Adaptation (QLoRA) on 947 synthetically generated maintenance reports generates a structured JSON report from that prompt. Retrieval-Augmented Fine-Tuning (RAFT) further grounds each recommendation in indexed maintenance procedures. Five ablation experiments, scored by BLEU-4, ROUGE-L, Hallucination Rate (HR), and an LLM-as-a-Judge rubric, compare the pipeline against a monolithic vision-language model (VLM) baseline and against partial configurations in which one component is removed. The complete system achieves BLEU-4 0.41, HR=4%, and Expert Score = 8.6/10 compared with 0.07, 65%, and 3.3/10 for the zero-shot VLM baseline. The QLoRA-adapted 1.5B model generates higher-quality reports than a 671B-parameter generalist API model given identical detection evidence, at 47 tokens per second on a single T4-class GPU. The results show that purpose-built decoupled architecture with a small domain-specific training corpus outperforms a generalist end-to-end model on this structured generation task.