Diffusion Large Language Models (DLLMs) offer a compelling alternative to Auto-Regressive models, but their deployment is constrained by high decoding cost. In this work, we identify a key inefficiency in DLLM decoding: while computation is parallelized over token blocks, only a small subset of tokens is decodable at each diffusion step, causing most compute to be wasted on non-decodable tokens. We further observe a strong correlation between attention-derived token importance and token-wise decoding probability. Based on this insight, we propose FOCUS -- an inference system designed for DLLMs. By dynamically focusing computation on decodable tokens and evicting non-decodable ones on-the-fly, FOCUS increases the effective batch size, alleviating compute limitations and enabling scalable throughput. Empirical evaluations demonstrate that FOCUS achieves up to 3.52$\times$ throughput improvement over the production-grade engine LMDeploy, while preserving or improving generation quality across multiple benchmarks. The FOCUS system is publicly available on GitHub: https://github.com/sands-lab/FOCUS.

翻译：扩散大语言模型（DLLMs）为自回归模型提供了一种引人注目的替代方案，但其部署受到高解码成本的限制。本研究揭示了DLLM解码过程中的一个关键低效问题：尽管计算在令牌块上并行化，但在每个扩散步骤中仅有少量令牌可被解码，导致大部分计算资源浪费在不可解码的令牌上。我们进一步观察到注意力机制衍生的令牌重要性与令牌级解码概率之间存在强相关性。基于这一发现，我们提出了FOCUS——一个专为DLLMs设计的推理系统。通过动态地将计算资源聚焦于可解码令牌并实时剔除不可解码令牌，FOCUS有效提升了批处理规模，缓解了计算限制并实现了可扩展的吞吐量。实证评估表明，FOCUS在保持或提升多项基准测试生成质量的同时，相较于生产级引擎LMDeploy实现了最高3.52倍的吞吐量提升。FOCUS系统已在GitHub开源：https://github.com/sands-lab/FOCUS。