mlsanityData to 7 Oct 2026

TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offload

arXiv 2605.20179 ↗First posted 19 May 2026cs.CLcited by 2

On the map: Mixture-of-experts → Vision mixture-of-experts

Abstract (arXiv)

Diffusion Large Language Models (dLLMs) have emerged as a competitive alternative to autoregressive (AR) models, offering better hardware utilization and bidirectional context through parallel block-level decoding. However, as dLLMs continue to scale up with mixture-of-experts (MoE) architectures, their deployment on resource-constrained devices remains an open challenge. Existing AR-based methods often incur either prohibitive I/O overhead or significant compute bottlenecks. In this work, we propose TIDE, a novel resource-efficient inference system that leverages the temporal stability of expert activations during the diffusion process within the block. Specifically, we leverage the temporal stability of expert activations during the diffusion process within the block and introduce an interval-based expert refresh strategy that updates the expert placement in an I/O-aware fashion. To ensure optimal performance, we formulate the inference scheduling as a mathematical programming problem, solving for the optimal interval that minimizes I/O traffic and CPU computation. Most importantly, TIDE is a lossless optimization that requires no model training, providing a "free lunch" acceleration for dLLM inference. In a single GPU-CPU system, we demonstrate that TIDE achieves up to 1.4× and 1.5× throughput improvements over prior baselines on LLaDA2.0-mini and LLaDA2.0-flash models, respectively.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts. Builds on 30 of its 36 references that are in the map. First posted on arXiv 19 May 2026.

Builds on

30 of 36 references.

References · 30

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2022Scalable Diffusion Models with Transformerscited by 5,931
  4. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  5. 2021Program Synthesis with Large Language Modelscited by 3,509
  6. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  7. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,014
  8. 2019HuggingFace's Transformers: State-of-the-art Natural Language Processingcited by 2,201
  9. 2025Large Language Diffusion Modelscited by 872
  10. 2023PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesiscited by 854
  11. 2024Open-Sora: Democratizing Efficient Video Production for Allcited by 637
  12. 2023FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPUcited by 567
18 more
  1. 2025Dream 7B: Diffusion Large Language Modelscited by 515
  2. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 404
  3. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Modelscited by 268
  4. 2022Post-training Quantization on Diffusion Modelscited by 262
  5. 2024MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Casescited by 216
  6. 2023PTQD: Accurate Post-Training Quantization for Diffusion Modelscited by 167
  7. 2025LLaDA2.0: Scaling Up Diffusion Language Models to 100Bcited by 154
  8. 2025dKV-Cache: The Cache for Diffusion Language Modelscited by 140
  9. 2025Accelerating Diffusion LLMs via Adaptive Parallel Decodingcited by 108
  10. 2023Fast Inference of Mixture-of-Experts Language Models with Offloadingcited by 88
  11. 2024Fiddler: CPU-GPU Orchestration for Fast Inference of Mixture-of-Experts Modelscited by 69
  12. 2025Diffusion Language Models Know the Answer Before Decodingcited by 68
  13. 2024ALISA: Accelerating Large Language Model Inference via Sparsity-Aware KV Cachingcited by 48
  14. 2025dInfer: An Efficient Inference Framework for Diffusion Language Modelscited by 33
  15. 2025Learning to Parallel: Accelerating Diffusion Large Language Models via Learnable Parallel Decodingcited by 26
  16. 2025Quant-dLLM: Post-Training Extreme Low-Bit Quantization for Diffusion Large Language Modelscited by 11
  17. 2025Taming the Memory Footprint Crisis: System Design for Production Diffusion LLM Servingcited by 5
  18. 2024Merino: Entropy-driven Design for Generative Language Models on IoT Devicescited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026CORA-Diff: Confidence-Oriented Residual Acceptance for Efficient Diffusion Language Model Inferencecited by 0
  2. 2026Denoising Surface: Modeling and Predicting Inference Cost for Diffusion LLM Servingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026TEAM: Temporal-Spatial Consistency Guided Expert Activation for MoE Diffusion Language Model Acceleration
  2. 2026OLED-MoE: Accelerating MoE-Based dLLM Inference via Inter-Iteration Locality-Aware Expert Offloading
  3. 2026LLaDA MoE v2: Scaling Mixture-of-Experts Diffusion Language Models
  4. 2026Dynamic Expert Sharing: Decoupling Memory from Parallelism in Mixture-of-Experts Diffusion LLMs
  5. 2026dMoE: dLLMs with Learnable Block Experts
  6. 2026REFLEX: Rethinking MoE Inference as Refinement-Aware Compute Allocation in Diffusion Language Models
  7. 2024Read-ME: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design
  8. 2026Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models