用随机游走门控混合多专家，提升网格分析性能
MME: Mixture of Mesh Experts with Random Walk Transformer Gating
- 设计随机游走门控机制，让专家专注擅长的物体类别
- 在网格分类、检索与语义分割任务上达最新水平
- 适合需要多类型网格分析的科研与工程应用
近年来，多种网格分析方法被提出，各自在不同物体类别上表现优异。本文提出一种新型混合专家（MoE）框架，融合多种方法的互补优势。设计新门控架构，促使每个专家专注于其擅长的类别。核心思想包括：(1) 在网格表面进行随机游走，有效捕捉各专家关注区域；(2) 采用注意力机制，使门控聚焦于对专家决策最有利的区域。为进一步提升性能，引入动态损失平衡策略，训练中调节多样性与相似性损失的权衡——多样性促进专家分化，相似性实现知识共享。该框架在网格分类、检索和语义分割任务上均取得当前最优结果。代码已开源：https://github.com/amirbelder/MME-Mixture-of-Mesh-Experts。
原文摘要 · Abstract (English)
In recent years, various methods have been proposed for mesh analysis, each offering distinct advantages and often excelling on different object classes. We present a novel Mixture of Experts (MoE) framework designed to harness the complementary strengths of these diverse approaches. We propose a new gate architecture that encourages each expert to specialise in the classes it excels in. Our design is guided by two key ideas: (1) random walks over the mesh surface effectively capture the regions that individual experts attend to, and (2) an attention mechanism that enables the gate to focus on the areas most informative for each expert's decision-making. To further enhance performance, we introduce a dynamic loss balancing scheme that adjusts a trade-off between diversity and similarity losses throughout the training, where diversity prompts expert specialization, and similarity enables knowledge sharing among the experts. Our framework achieves state-of-the-art results in mesh classification, retrieval, and semantic segmentation tasks. Our code is available at: https://github.com/amirbelder/MME-Mixture-of-Mesh-Experts.
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。