发表机构
Hefei University of Technology; Universiti Malaya（合肥工业大学; 马来亚大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对跨领域PCBA视觉问答，提出任务感知GRPO框架，整合多组件语义奖励与距离感知奖励，结合推理校正，在官方挑战中取得83.24分。
AI 中文摘要
在自动化印刷电路板组装（PCBA）检测中，基于标准的决策要求系统联合推理细粒度视觉线索、元件语义和制造知识。尽管大型视觉语言模型（VLM）提供了有前景的基础，但其部署受到标准衍生样本与真实生产线图像之间领域差异的阻碍，同时输出空间异构，涵盖基于选择和数值计数任务。为应对这些挑战，我们提出了一种用于跨领域PCBA视觉问答的多模态推理框架。该框架将标准衍生、真实世界及辅助PCB领域数据转换为统一的指令格式，并构建与视觉证据、问题语义、候选选项和真实答案对齐的验证推理轨迹。我们进一步引入了任务感知组相对策略优化（GRPO），它超越精确匹配监督，通过整合基于选择问题的多组件语义奖励、计数问题的距离感知奖励，以及用于有效输出的辅助格式奖励。在推理阶段，结合答案选项语义一致性校正、自一致性投票和多模型仲裁以提高预测鲁棒性。所提系统在官方PCBA标准到真实大挑战排行榜上取得了83.24的总体得分，证明了任务感知奖励设计和鲁棒推理在跨领域PCBA视觉问答中的有效性。
英文摘要
In automated Printed Circuit Board Assembly (PCBA) inspection, standards-guided decisions require systems to jointly reason over fine-grained visual cues, component semantics, and manufacturing knowledge. Although large vision-language models (VLMs) provide a promising foundation, their deployment is hindered by the domain shift between standards-derived samples and real-world production-line imagery, together with heterogeneous output spaces spanning choice-based and numerical counting tasks. To address these challenges, we propose a multimodal reasoning framework for cross-domain PCBA visual question answering. The framework converts standards-derived, real-world, and auxiliary PCB-domain data into a unified instruction format and constructs verified reasoning traces aligned with visual evidence, question semantics, candidate options, and ground-truth answers. We further introduce Task-Aware Group Relative Policy Optimization (GRPO), which moves beyond exact-match supervision by integrating multi-component semantic rewards for choice-based questions, distance-aware rewards for counting questions, and an auxiliary format reward for valid outputs. During inference, answer-option semantic consistency correction, self-consistency voting, and multi-model arbitration are combined to improve prediction robustness. The proposed system achieves an Overall Score of 83.24 on the official PCBA Standard-to-Real Grand Challenge leaderboard, demonstrating the effectiveness of task-aware reward design and robust inference for cross-domain PCBA visual question answering.
Comments8 pages, 2 figures. Accepted to the 34th ACM International Conference on Multimedia (ACM MM 2026)
