Jianru Xu — accepted AI-conference papers

2025

Accelerating Parallel Diffusion Model Serving with Residual Compression

Jiajun Luo, Yicheng Xiao, Jianru Xu, Yangxiu You, Rongwei Lu, Chen Tang +2

NeurIPS 2025poster

Diffusion models produce realistic images and videos but require substantial computational resources, necessitating multi-accelerator parallelism for real-time deployment. However, parallel inference introduces significant communication overhead from exchanging large activations between devices, lim…

Cited by 0 Source ↗code ↗Scholar ↗

2025

DICE: Staleness-Centric Optimizations for Parallel Diffusion MoE Inference

Jiajun Luo, Lizhuo Luo, Jianru Xu, Jiajun Song, Rongwei Lu, Chen Tang +1

ICCV 2025poster

Mixture-of-Experts-based (MoE-based) diffusion models demonstrate remarkable scalability in high-fidelity image generation, yet their reliance on expert parallelism introduces critical communication bottlenecks. State-of-the-art methods alleviate such overhead in parallel diffusion inference through…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗