activity
20242026
collaborators

9 papers

cs.DC2026

CommBench: Can LLMs Write Correct and Efficient GPU Communication Code?

Shuang Ma, Yuyi Li, Yihan Zhang +12

Training and serving large language models (LLMs) rely heavily on high-performance GPU communication, yet implementing efficient GPU communication primitives requires deep expertis…

cs.CV2026

SVG-EAR: Parameter-Free Linear Compensation for Sparse Video Generation via Error-aware Routing

Xuanyi Zhou, Qiuyang Mang, Shuo Yang +7

Diffusion Transformers (DiTs) have become a leading backbone for video generation, yet their quadratic attention cost remains a major bottleneck. Sparse attention reduces this cost…

cs.DC2026

Flash-KMeans: Fast and Memory-Efficient Exact K-Means

Shuo Yang, Haocheng Xi, Yilong Zhao +10

-means has historically been positioned primarily as an offline processing primitive, typically used for dataset organization or embedding preprocessing rather than as a first-c…

cs.DC2026

UCCL-EP: Portable Expert-Parallel Communication

Ziming Mao, Yihan Zhang, Chihan Cui +9

Mixture-of-Experts (MoE) workloads rely on expert parallelism (EP) to achieve high GPU efficiency. State-of-the-art EP communication systems such as DeepEP demonstrate strong perfo…

cs.DC2026

SkyNomad: On Using Multi-Region Spot Instances to Minimize AI Batch Job Cost

Zhifei Li, Tian Xia, Ziming Mao +9

AI batch jobs such as model training, inference pipelines, and data analytics require substantial GPU resources and often need to finish before a deadline. Spot instances offer 3-1…

cs.DC2025

SkyWalker: A Locality-Aware Cross-Region Load Balancer for LLM Inference

Tian Xia, Ziming Mao, Jamison Kerney +5

Serving Large Language Models (LLMs) efficiently in multi-region setups remains a challenge. Due to cost and GPU availability concerns, providers typically deploy LLMs in multiple…