OPERA-MoE: Output-aware Routing Perturbation Alignment via Observable Discrepancy for MoE Quantization
Abstract
AbstractMixture-of-Experts (MoE) architectures have become a dominant paradigm for scaling large language models (LLMs), but their large memory footprint still makes post-training quantization (PTQ) essential for practical deployment. Unlike dense LLMs, quantized MoE models exhibit routing perturbations, where quantization alters expert routing behavior. Existing router alignment methods mitigate this issue by uniformly minimizing routing discrepancies across all tokens. However, we observe that routing perturbations exhibit highly heterogeneous relationships with observable output discrepancy: many tokens preserve nearly identical output distributions despite substantial routing changes. Motivated by this observation, we propose OPERA, an output-aware router alignment strategy for quantized MoE models that selectively prioritizes tokens exhibiting observable output discrepancy after quantization. We further analyze output-preserving routing perturbations and show that they frequently occur on highly predictable tokens, where substantial routing perturbations and hidden-state deviations may not strongly affect projected output distributions. Experiments across multiple MoE LLMs demonstrate that OPERA generally improves over conventional router alignment, with particularly noticeable gains under aggressive low-bit quantization settings.