activity
20242026
collaborators

7 papers

cs.DC2026

GRACE-MoE: Grouping and Replication with Locality-Aware Routing for Efficient Distributed MoE Inference

Yu Han, Lehan Pan, Jie Peng +4

Sparse Mixture of Experts (SMoE) enables scalable parameter growth in large language models (LLMs) by selectively activating a subset of experts, and its large parameter count nece…

cs.CL2026

Making Every Verified Token Count: Adaptive Verification for MoE Speculative Decoding

Lehan Pan, Ziyang Tao, Ruoyu Pang +3

Tree-based speculative decoding accelerates autoregressive generation by verifying multiple draft candidates in parallel, but this advantage weakens for sparse Mixture-of-Experts (…

cs.LG2026

DepCap: Adaptive Block-Wise Parallel Decoding for Efficient Diffusion LM Inference

Xiang Xia, Wuyang Zhang, Jiazheng Liu +2

Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive language generation due to their potential for parallel decoding and global refinement of…

cs.LG2026

Understanding the Role of Hallucination in Reinforcement Post-Training of Multimodal Reasoning Models

Gengwei Zhang, Jie Peng, Zhen Tan +6

The recent success of reinforcement learning (RL) in large reasoning models has inspired the growing adoption of RL for post-training Multimodal Large Language Models (MLLMs) to en…

cs.CR2025

LightDefense: A Lightweight Uncertainty-Driven Defense against Jailbreaks via Shifted Token Distribution

Zhuoran Yang, Yanyong Zhang

Large Language Models (LLMs) face threats from jailbreak prompts. Existing methods for defending against jailbreak attacks are primarily based on auxiliary models. These strategies…

cs.AR2025

T-MAN: Enabling End-to-End Low-Bit LLM Inference on NPUs via Unified Table Lookup

Jianyu Wei, Qingtao Li, Shijie Cao +5

Large language models (LLMs) are increasingly deployed on customer devices. To support them, current devices are adopting SoCs (System on Chip) with NPUs (Neural Processing Unit) i…