Hugging Face Trending Papers

BlockServe: Block-Grained Continuous Batching for High-Throughput Diffusion LLM Serving

Read the original on Hugging Face Trending Papers →

Efficient serving of diffusion large language models (dLLMs) is hindered by convergence heterogeneity: when batching multiple requests, different sequences converge at different rates, causing faster requests to stall behind slower stragglers and introducing compute bubbles and tail latency. We present BlockServe, a continuous batching framework that integrates block-grained scheduling -- immediately evicting completed requests at block boundaries -- with mixed-state execution that extends dual cache and parallel decoding to heterogeneous batches via gather-scatter indexing.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.

arXiv Machine Learning
Jul 13

BlockServe: Block-Grained Continuous Batching for High-Throughput Diffusion LLM Serving

arXiv:2607. 08930v1 Announce Type: new Abstract: Efficient serving of diffusion large language models (dLLMs) is hindered by convergence heterogeneity: when batching multiple requests, different sequences converge at different rates, causing faster requests to stall behind slower stragglers and introducing compute bubbles and tail latency.

By Yuanjie Zhu, Liangwei Yang, Ke Xu, Weizhi Zhang, Shanghao Li, Zihe Song, Philip S. Yu
Hugging Face Trending Papers
Jul 20

FlowBlock: Wavefront-Parallel Decoding for Self-Correcting Diffusion Language Models

Block-wise diffusion large language models (dLLMs) decode sequentially at the block level, enabling effective KV-cache reuse across blocks but making inter-block decoding strictly serial. Prior work has attempted to unlock inter-block parallelism through post-training methods, but achieves only modest speedups and often degrades accuracy.