@article{flashdllmioawarekvcachingandparalleldeco,
  title = {Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs},
  author = {Quan Nguyen-Tri and Mukul Ranjan and Zhiqiang Shen},
  year = {2026},
  eprint = {2609.26796},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.26796},
}