2026
Mixture-of-Trees: Learning to Select and Weigh Reasoning Paths for Efficient LLM Inference
AAAI 2026technical
We introduce Mixture-of-Trees (MoT), a novel framework that integrates sparse expert activation with structured tree-based reasoning for efficient LLM inference. MoT employs a learned gating mechanism to selectively activate only the most relevant expert reasoning trees for each problem, where exper