collaborators

9 papers

cs.CL2026

When Does Divide and Conquer Work for Long Context LLM? A Noise Decomposition Framework

Zhen Xu, Shang Zhu, Jue Wang +5

We investigate the challenge of applying Large Language Models (LLMs) to long texts. We propose a theoretical framework that distinguishes the failure modes of long context tasks i…

cs.LG2025

Kitty: Accurate and Efficient 2-bit KV Cache Quantization with Dynamic Channel-wise Precision Boost

Haojun Xia, Xiaoxia Wu, Jisen Li +12

The KV cache is a dominant memory bottleneck for LLM inference. While 4-bit KV quantization preserves accuracy, 2-bit often degrades it, especially on long-context reasoning. We cl…

cs.LG2025

Beat the long tail: Distribution-Aware Speculative Decoding for RL Training

Zelei Shao, Vikranth Srivatsa, Sanjana Srivastava +12

Reinforcement learning(RL) post-training has become essential for aligning large language models (LLMs), yet its efficiency is increasingly constrained by the rollout phase, where…

cs.LG2025

Ladder-residual: parallelism-aware architecture for accelerating large model inference with communication overlapping

Muru Zhang, Mayank Mishra, Zhongzhu Zhou +7

Large language model inference is both memory-intensive and time-consuming, often requiring distributed algorithms to efficiently scale. Various model parallelism strategies are us…

cs.LG2025

FloE: On-the-Fly MoE Inference on Memory-constrained GPU

Yuxin Zhou, Zheng Li, Jun Zhang +5

With the widespread adoption of Mixture-of-Experts (MoE) models, there is a growing demand for efficient inference on memory-constrained devices. While offloading expert parameters…

cs.LG2025

HMI: Hierarchical Knowledge Management for Efficient Multi-Tenant Inference in Pretrained Language Models

Jun Zhang, Jue Wang, Huan Li +6

The significant computational demands of pretrained language models (PLMs), which often require dedicated hardware, present a substantial challenge in serving them efficiently, esp…