Author

Zhengzhong Tu

View on Semantic Scholar

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Jul 2026

FlashDiff: Efficient Regional Execution and Scheduling for Diffusion Model Serving

FlashDiff is a diffusion serving system that improves inference efficiency through adaptive regional execution and scheduling that reduces end-to-end serving latency by 30-97% and improves throughput by 1.2-2.2x.

Yaqi Qiao, Ping He, Songru Xie et al. · arXiv.org · 0 citations