Model WarsJuly 6, 2026via Apple Machine Learning

Path-Constrained Mixture-of-Experts

Why it matters

Apple researchers have identified a fundamental inefficiency in how Mixture-of-Experts models route tokens across layers, proposing path-constrained architectures that could significantly reduce compute waste and improve inference efficiency—a key competitive lever in the race to optimize LLM deployment.

Key signals

  • N^L possible expert paths in standard MoE (N experts, L layers) remain mostly unexplored
  • Tokens cluster into small fraction of paths aligned with linguistic function
  • Vast majority of paths statistically unused, revealing architectural inefficiency
  • Path-constrained MoE architectures proposed as optimization solution
  • Published by Apple Machine Learning Research

The hook

Apple just revealed why your MoE models are wasting 99% of their paths. Here's what that means for inference cost.

Sparse Mixture-of-Experts (MoE) architectures route each token through a subset of experts at each layer independently. We propose viewing MoE computation through the lens of expert paths—the sequence of expert selections a token makes across all layers. This perspective reveals that, despite N^L possible paths for N experts across L layers, tokens in practice cluster into a small fraction of paths that align with linguistic function, yet the vast majority of paths remain unexplored, representing a statistical inefficiency. This motivates architectures that constrain the effective path space…

Get stories like this every Friday.

The 5 AI stories that matter — free, in your inbox.

Free forever. No spam.