cs.LGNov 16, 2026

The Hidden Cost of MoE Diffusion Decoding: Profiling and Fixing the Sampler

Authors: Harshit Samrat, Ramakrishna Paramahamsa Kompella

Organizations: Indian Institute of Technology Madras

Abstract

Masked diffusion language models (dLLMs) decode by iteratively denoising a full canvas of tokens, so every forward pass runs a per-token sampling and stopping pipeline over the entire canvas-by-vocabulary logits tensor. We profile single-user decoding of a 26B-parameter diffusion Mixture-of-Experts model on a bandwidth-bound edge accelerator (NVIDIA GB10) and report a result that is easy to miss: after weight quantisation, this logits pipeline, not the expert GEMMs, is the single largest consumer of decode time. Unlike autoregressive decoding, where sampling touches one position's logits per committed token, diffusion decoding processes S times as many logit entries per output token, where S is the number of denoising steps. We show this pipeline is largely redundant or fusable and remove it in two lossless tiers: a bit-exact tier whose output is byte-identical to the reference decoder, and an accuracy-parity tier using a fused single-pass entropy kernel and Gumbel-max sampling. Combined with a W4A16 expert kernel, the full stack reaches 2.4-2.7 times end-to-end speedup over bf16 at unchanged benchmark accuracy on GSM8K, HumanEval, and MATH, on a single 273 GB/s device. We confirm the finding on a second, independently built diffusion MoE (LLaDA2.0-mini: pipeline 28.8% of decode, fused W4A16 3.3 times at accuracy parity) and show it disappears on a dense diffusion model (3.6%). The cost is specific to MoE diffusion, where quantisation makes the backbone cheap. Code and quantised weights are released.

Figures & tables

Explore similar work

Sep 29, 2026

E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models

Masked diffusion models (MDMs) generate sequences by progressively unmasking several tokens per denoising step, but their reverse process is typically factorized over positions, limiting sample quality in the few-step regime where diffusion's speed advantage over autoregressive decoding matters most. A recent line of work introduces a continuous Gaussian latent, trained as a variational autoencoder, to capture correlations across positions, but such approaches are prone to posterior collapse, where the latent is silently ignored. We propose Enhanced Mixture-of-Experts (E-MoE), which builds the reverse process as a mixture of factorized distributions over a discrete shared latent given by the expert-routing decisions of a Mixture-of-Experts (MoE) backbone, without increasing active parameters over the factorized baseline. Across synthetic multi-modal benchmarks, binarized MNIST, and LM1B, E-MoE improves few-step generation over factorized baselines.
Mar 9, 2026

DyLLM: Efficient Diffusion LLM Inference via Saliency-based Token Selection and Partial Attention

Masked diffusion language models enable parallel token decoding, providing a promising alternative to the sequential nature of autoregressive generation. However, their iterative denoising process remains computationally expensive because it repeatedly processes the entire sequence at every step. We observe that across these diffusion steps, most token representations remain stable; only a small subset, which we term salient tokens, contributes meaningfully to the next update. Leveraging this temporal sparsity, we present DyLLM, a training-free inference framework that accelerates decoding by selectively computing only these salient tokens. DyLLM identifies saliency by measuring the cosine similarity of attention contexts between adjacent denoising steps. It recomputes feed-forward and attention operations only for salient tokens while reusing cached activations for the remainder. Across diverse reasoning and code-generation benchmarks, DyLLM achieves up to 9.6x higher throughput while largely preserving the baseline accuracy of representative open-source diffusion LLMs, LLaDA, and Dream.
May 29, 2026

dMoE: dLLMs with Learnable Block Experts

Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive models, offering competitive performance while naturally supporting parallel decoding. However, as dLLMs are increasingly integrated with Mixture-of-Experts (MoE) architectures to scale model capacity, a fundamental mismatch arises between block parallel decoding and token-level expert selection. Specifically, each dLLM forward pass processes multiple tokens with bidirectional dependencies, whereas conventional MoE layers route each token independently. This mismatch substantially increases the number of uniquely activated experts, making inference increasingly memory-bound. To address this, we propose dMoE, a simple yet effective block-level MoE framework. The central idea of dMoE is to aggregate token-level expert distributions within each block into a unified block-level expert distribution, which is then used to guide block expert routing in a coherent manner. In this way, dMoE substantially reduces the number of uniquely activated experts during inference without sacrificing performance, thereby mitigating the memory-bound bottleneck. Extensive experiments across a variety of benchmarks demonstrate the effectiveness of dMoE. On average, dMoE reduces the number of uniquely activated experts from 69.5 to 16.6 while retaining 99.55% of the original performance. Meanwhile, it reduces memory usage by 74.36% to 76.78% and achieves 1.36×1.36\times to 2.04×2.04\times end-to-end latency speedup. Code is available at: https://github.com/fscdc/dMoE.