AI RESEARCH
DBES: A Systematic Benchmark and Metric Suite for Evaluating Expert Specialization in Large-Scale MoEs
arXiv CS.LG
•
ArXi:2605.18498v1 Announce Type: new Expert specialization in Mixture-of-Experts (MoE) models remains poorly understood, with traditional evaluations conflating architectural load-balancing with functional specialization. We Critical findings nstrate distinct specialization paradigms across models: Qwen-series exhibit modular specialization with high domain isolation, while DeepSeek and GLM employ distributed collaboration. However, we emphasize that specialization is a diagnostic dimension, necessary but not sufficient for downstream performance.