PCoMoE introduces a path‑compositional execution framework that moves Mixture‑of‑Experts inference from coarse‑grained expert selection to fine‑grained path composition. It uses a path‑level formulation of expert computation, a compatibility‑aware layer‑wise pruning strategy to eliminate low‑value path combinations, and a hardware‑friendly execution engine that reuses sub‑expert structures with bounded overhead. Experiments show up to a 1.31× speedup and a 10% accuracy improvement over existing MoE inference methods.
Mixture-of-Experts (MoE) architectures scale Large Language Model (LLM) capacity efficiently by activating a sparse subset of experts per token. However, modern MoE inference remains heavily constrain...
arXiv:2602. 06154v2 Announce Type: replace Abstract: Mixture-of-Experts (MoE) models scale large language models efficiently by sparsely activating experts, but once an expert is selected, it is executed fully.
arXiv:2606. 01062v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models have become a leading approach for decoupling parameter count from computational cost in large language models, yet effectively scaling MoE performance remains a challenge.
arXiv:2606. 15453v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) based large language models (LLMs), such as Qwen and DeepSeek, have recently emerged as an effective approach to improving model capacity without proportionally increasing computational cost.
The paper introduces Ban&Pick, a post‑training, plug‑and‑play routing strategy for Sparse Mixture‑of‑Experts large language models. It identifies and reinforces a small group of highly influential experts while dynamically pruning redundant ones, leading to accuracy gains across math, code, and reasoning benchmarks. Experiments on DeepSeek and Qwen3 show notable performance improvements and a 1.25× inference speedup without retraining or architectural changes.
arXiv:2511. 04805v2 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) models have shown strong potential in scaling language models efficiently by activating only a small subset of experts per input.
arXiv:2503. 05641v4 Announce Type: replace-cross Abstract: Combining existing pre-trained LLMs is a promising approach for diverse reasoning tasks.
arXiv:2608.21614v1 Announce Type: new Abstract: Chain-of-thought (CoT) prompting improves LLM reasoning by decomposing complex problems into intermediate steps, but its sequential nature increases de...
ExFold is a training‑free expert‑folding framework that jointly accelerates the prefill and decode phases of Mixture‑of‑Experts (MoE) models by projecting the contributions of excluded experts onto a retained expert set using calibrated scalar projectors. It treats both phases as a budgeted output‑approximation problem, achieving token‑level Top‑K folding for prefill and batch‑level expert‑pool folding for decode. Implemented as a plug‑and‑play plugin in vLLM with a lightweight CUDA kernel, ExFold delivers up to 1.41× TTFT and 2.45× TPOT speedups while preserving about 99% of the original model quality.
arXiv:2606. 16825v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures efficiently scale Large Language Models (LLMs) by activating only a small fraction of their experts per token, yet the full parameter count - dominated by the expert parameters - must be held in training and inference memory.
arXiv:2609.25809v1 Announce Type: new Abstract: Fine-grained mixture-of-experts (MoE) architectures have become a mainstream design for open-weight LLMs, with hundreds of experts and increasingly man...
arXiv:2606. 05538v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models achieve strong performance through conditional computation, but their large parameter footprint poses deployment challenges.