A training-free acceleration method for diffusion language models that achieves a 4x speedup in image generation.
March 17, 2026
Original Paper
LADR: Locality-Aware Dynamic Rescue for Efficient Text-to-Image Generation with Diffusion Large Language Models
arXiv · 2603.13450
The Takeaway
LADR exploits the spatial Markov property of images to prioritize decoding at 'generation frontiers.' Unlike other acceleration techniques, it requires no retraining and maintains or improves spatial reasoning and fidelity in multimodal generation tasks.
From the abstract
Discrete Diffusion Language Models have emerged as a compelling paradigm for unified multimodal generation, yet their deployment is hindered by high inference latency arising from iterative decoding. Existing acceleration strategies often require expensive re-training or fail to leverage the 2D spatial redundancy inherent in visual data. To address this, we propose Locality-Aware Dynamic Rescue (LADR), a training-free method that expedites inference by exploiting the spatial Markov property of i