9 papers
X-OmniClaw Technical Report: A Unified Mobile Agent for Multimodal Understanding and Interaction
Xiaoming Ren, Ru Zhen, Chao Li +11
Inspired by the development of OpenClaw, there is a growing demand for mobile-based personal agents capable of handling complex and intuitive interactions. In this technical report…
VisMMOE: Exploiting Visual-Expert Affinity for Efficient Visual-Language MoE Offloading
Cheng Xu, Xiaofeng Hou, Jiacheng Liu +1
Large-scale vision-language mixture-of-experts (VL-MoE) models provide strong multimodal capability, but efficient deployment on memory-constrained platforms remains difficult. Exi…
CODO: An Automated Compiler for Comprehensive Dataflow Optimization
Weichuang Zhang, Yiquan Wang, Xinzhou Zhang +6
FPGAs are well-suited for dataflow architectures that process data in a streaming or pipelined manner, thus satisfying the high computational and communication demands of emerging…
PCR: A Prefetch-Enhanced Cache Reuse System for Low-Latency RAG Serving
Wenfeng Wang, Xiaofeng Hou, Peng Tang +5
Retrieval-Augmented Generation (RAG) systems enhance the performance of large language models (LLMs) by incorporating supplementary retrieved documents, enabling more accurate and…
MoE-SpeQ: Speculative Quantized Decoding with Proactive Expert Prefetching and Offloading for Mixture-of-Experts
Wenfeng Wang, Jiacheng Liu, Xiaofeng Hou +5
The immense memory requirements of state-of-the-art Mixture-of-Experts (MoE) models present a significant challenge for inference, often exceeding the capacity of a single accelera…
MoE-Prism: Disentangling Monolithic Experts for Elastic MoE Services via Model-System Co-Designs
Xinfeng Xia, Jiacheng Liu, Xiaofeng Hou +5
Mixture-of-Experts (MoE) scales model capacity through sparse activation, and is becoming an important architecture for large language models (LLMs). However, existing MoE serving…