用答案语义引导专家选择，提升视觉问答的推理一致性与灵活性。
CoGR-MoE: Concept-Guided Expert Routing with Consistent Selection and Flexible Reasoning for Visual Question Answering
- 基于答案语义设计专家路由，训练时引导更合理的专家分配。
- 通过重加权选中专家，生成区分度高的选项表示，提升判断精度。
- 适合需要精准选项推理的视觉问答任务，尤其在复杂场景下表现优。
视觉问答（VQA）要求模型根据视觉和文本证据选择正确答案。现有混合专家（MoE）方法通过概念分组或基于样本路由改进选项推理，但路由不稳定会导致相同问题类型下专家选择不一致，而过度稳定又会降低灵活性。为此，我们提出概念引导路由框架（CoGR-MoE），在训练阶段引入答案选项的语义信息以指导专家选择。随后，利用选项特征重新加权已选专家，生成每个候选选项的判别性表征，并通过对比学习优化这些表征。实验表明，CoGR-MoE在多个VQA任务上均取得优异性能，验证了该方法的有效性。
原文摘要 · Abstract (English)
Visual Question Answering (VQA) requires models to identify the correct answer options based on both visual and textual evidence. Recent Mixture-of-Experts (MoE) methods improve option reasoning by grouping similar concepts or routing based on examples. However, unstable routing can lead to inconsistent expert selection in the same question type, while overly stable routing may reduce flexibility. To address this, we propose Concept-Guided Routing framework (CoGR-MoE), which incorporates semantics of the answer options to guide expert selection in the training phase. Next, option features are used to reweight the selected experts, producing discriminative representations for each candidate option. These option-level representations are further used for option comparison and optimized via contrastive learning. The experimental results indicate that CoGR-MoE delivers strong performance across multiple VQA tasks, demonstrating the effectiveness of our approach.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。