activity
20242026
collaborators

8 papers

cs.DC2026

KernelFlume: Elastic Core-Attention Scaling for Agentic Long-Context Decoding

Guangyu Xiang, Xueze Kang, Lin Zhang +4

LLM serving is increasingly dominated by long and dynamic decode workloads from agents, reasoning models, and extended conversations. When bursty long-context demand exceeds deploy…

cs.CL2026

AGoQ: Activation and Gradient Quantization for Memory-Efficient Distributed Training of LLMs

Wenxiang Lin, Juntao Huang, Luhan Zhang +5

Quantization is a key method for reducing the GPU memory requirement of training large language models (LLMs). Yet, current approaches are ineffective for 4-bit activations and 8-b…

cs.DC2026

ZipCCL: Efficient Lossless Data Compression of Communication Collectives for Accelerating LLM Training

Wenxiang Lin, Xinglin Pan, Ruibo Fan +2

Communication has emerged as a critical bottleneck in the distributed training of large language models (LLMs). While numerous approaches have been proposed to reduce communication…

cs.DC2025

Efficient MoE Inference with Fine-Grained Scheduling of Disaggregated Expert Parallelism

Xinglin Pan, Shaohuai Shi, Wenxiang Lin +4

The mixture-of-experts (MoE) architecture scales model size with sublinear computational increase but suffers from memory-intensive inference due to KV caches and sparse expert act…

cs.DC2025

HierMoE: Accelerating MoE Training with Hierarchical Token Deduplication and Expert Swap

Wenxiang Lin, Xinglin Pan, Lin Zhang +3

The sparsely activated mixture-of-experts (MoE) transformer has become a common architecture for large language models (LLMs) due to its sparsity, which requires fewer computationa…

cs.LG2025

FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts Models

Xinglin Pan, Wenxiang Lin, Lin Zhang +5

Recent large language models (LLMs) have tended to leverage sparsity to reduce computations, employing the sparsely activated mixture-of-experts (MoE) technique. MoE introduces fou…