Algorithm/Software

Cachemir: Fully Homomorphic Encrypted Inference of Generative Large Language Model with KV Cache
一种在全同态加密大语言模型推理中高效集成KV缓存以加速自回归解码的框架。
Ye Yu, Yifan Zhou, Yi Chen, Pedro Soto, Wenjie Xiong, 李萌
论文
Jetson-PI: Towards Onboard Real-Time Robot Control via Foresight-Aligned Asynchronous Inference
通过前瞻对齐异步推理与优化的边缘运行时，实现视觉-语言-动作模型的实时机载部署。
Zebin Yang, Qi Wang, Yunhe Wang, Xiurui Guo, Bo Yu, Shaoshan Liu, Jiafeng Xu, Hao Dong, 李萌
论文 代码 项目
ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation
一种采用混合CKKS/TFHE计算的高效、准确隐私保护大语言模型自回归解码框架。
Jiangrui Yu, Baosheng Zhang, Liang Kong, Lin Ding, Yi Chen, Ye Yu, Mingzhe Zhang, 李萌
论文
MatMoE: Matryoshka Mixture-of-Experts with Dynamic Mixed-Precision Quantization for Efficient Inference
Renjie Wei, Haochen Huang, Chuyu Qiu, Jinqi Wen, Dehao Xu, 李萌
Helios: Melting Kernel Boundaries for GPU-Accelerated HE via Graph Rewriting and Microarchitecture-Aware Mapping
Yi Chen, Ziyu Tang, Chao Yang, Guang Fan, Mingzhe Zhang, 李萌
RePart: Efficient Hypergraph Partitioning with Logic Replication Optimization for Multi-FPGA System
Zizhuo Fu, Yifan Zhou, Zhaoxin Lu, Guangyu Sun, Runsheng Wang, 李萌, Yibo Lin
论文
Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse
Zizhuo Fu, Wenxuan Zeng, Runsheng Wang, 李萌
论文
HyPER: Bridging Exploration and Exploitation for Scalable LLM Reasoning with Hypothesis Path Expansion and Reduction
论文
TEAM: Temporal–Spatial Consistency Guided Expert Activation for MoE Diffusion Language Model Acceleration
Linye Wei, Zixiang Luo, Pingzhi Tang, 李萌
论文
Breaking the Reward Barrier: Accelerating Tree-of-Thought Reasoning via Speculative Exploration
A speculative exploration system that breaks reward synchronization barriers to accelerate scalable Tree-of-Thought reasoning.
Shuzhang Zhong, Haochen Huang, Shengxuan Qiu, Pengfei Zuo, Runsheng Wang, 李萌
论文