dLLM Notes

Paper notes · ICML 2026

dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching

Zhiyuan Liu, Yicun Yang, Yaojie Zhang, Junjie Chen, Chang Zou, Qingyan Wei, Shaobo Wang, Yichen Zhu, Linfeng Zhang

Caches more than keys and values: per layer it stores K, V, attention output and FFN output, separately for the prompt and the response. The prompt cache is refreshed every KpK_p steps (long), the response cache fully every KrK_r steps (short), and in between only the response tokens whose value vectors changed most are recomputed.

The PDF's license does not allow reposting it here; page and line numbers refer to the version linked above.