195 citations · 208 across the 31 of their papers we have counts for
36 papers
DScale: Scaling Block-Diffusion Speculative Decoding with Adaptive Verification
Rongjian Chen, Minxian Xu, Zhengxin Fang +2
Growing large language model applications demand efficient inference. At high concurrency, block-diffusion speculative decoding suffers from verification padding, rejected candidat…
Memory in the Sky: Low-Altitude Question Answering with Multi-Agent Memory Aggregation
Chengyang Li, Yujie Wan, Shuai Wang +6
This paper studies low-altitude question answering (LAQA), in which distributed unmanned aerial vehicle (UAV) memories are aggregated at a ground server to answer questions about o…
AgentLoop: Runtime Control of Slot-closed Execution Loops for Tool-augmented LLM Agents
Wanyi Zheng, Minxian Xu, Kan Hu +2
Tool-augmented large language model (LLM) agents are becoming an important execution unit in service computing, but existing agent loops still lack explicit runtime signals for ass…
CoCoScale: Leveraging Layer-wise Scaling to Unlock the Potential of Online LLM Serving
Jingfeng Wu, Yiyuan He, Minxian Xu +7
Online large language model (LLM) serving has become the backbone of modern AI applications, powering diverse downstream services through shared hardware clusters. However, modern…
BrownoutMoE: Structure-Aware Expert Grouping for Efficient and Accurate LLM Web-based Services
Yi Ding, Minxian Xu, Zhengxin Fang +2
Mixture-of-Experts (MoE) large language models (LLMs) are increasingly deployed in Web-facing services, where inference must be both accurate and responsive under bursty demand. Al…
RolloutPipe: Overlapping Pipelined Rollout and Training in Disaggregated On-Policy LLM Reinforcement Learning
Rongjian Chen, Jianmin Hu, Kejiang Ye +1
Large language model (LLM) post-training for reasoning increasingly relies on reinforcement learning with verifiable rewards (RLVR), where models learn from ground-truth feedback o…
SwiftCache: Efficient LLM Serving for Multi-turn Conversations with Heterogeneous KV Cache Sharing
Jianmin Hu, Minxian Xu, Sa Wang +5
Multi-turn conversation is a fundamental scenario in LLM applications, widely used in chatbots and AI agents. As the conversation evolves, historical tokens accumulate continuously…
Cloud-native and Distributed Systems for Efficient and Scalable Large Language Models -- A Research Agenda
Minxian Xu, Jingfeng Wu, Shengye Song +16
The rapid rise of Large Language Models (LLMs) has revolutionized various artificial intelligence (AI) applications, from natural language processing to code generation. However, t…
Memory Centric Power Allocation for Multi-Agent Embodied Question Answering
Chengyang Li, Shuai Wang, Kejiang Ye +5
This paper considers multi-agent embodied question answering (MA-EQA), which enables robot teams to answer queries based on their long-horizon observations. In contrast to existing…
DOPD: A Dynamic PD-Disaggregation Architecture for Maximizing Goodput in LLM Inference Serving
Junhan Liao, Minxian Xu, Wanyi Zheng +4
To meet strict Service-Level Objectives (SLOs),contemporary Large Language Models (LLMs) decouple the prefill and decoding stages and place them on separate GPUs to mitigate the di…
BanaServe: Unified KV Cache and Dynamic Module Migration for Balancing Disaggregated LLM Serving in AI Infrastructure
Yiyuan He, Minxian Xu, Jingfeng Wu +7
Large language models (LLMs) are increasingly deployed in AI infrastructure, driving the need for high throughput, resource efficient serving systems. Disaggregated LLM serving, wh…
Edge Collaborative Gaussian Splatting with Integrated Rendering and Communication
Yujie Wan, Chenxuan Liu, Shuai Wang +5
Gaussian splatting (GS) struggles with degraded rendering quality on low-cost devices. To address this issue, we present edge collaborative GS (ECO-GS), where each user can switch…
TD3-Sched: Learning to Orchestrate Container-based Cloud-Edge Resources via Distributed Reinforcement Learning
Shengye Song, Minxian Xu, Kan Hu +2
Resource scheduling in cloud-edge systems is challenging as edge nodes run latency-sensitive workloads under tight resource constraints, while existing centralized schedulers can s…
Communication Efficient Robotic Mixed Reality with Gaussian Splatting Cross-Layer Optimization
Chenxuan Liu, He Li, Zongze Li +5
Realizing low-cost communication in robotic mixed reality (RoboMR) systems presents a challenge, due to the necessity of uploading high-resolution images through wireless channels.…
Joint Beamforming Design for RIS-Empowered NOMA-ISAC Systems
Chunjie Wang, Xuhui Zhang, Jinke Ren +6
This paper investigates a reconfigurable intelligent surface (RIS)-assisted integrated sensing and communication (ISAC) system and proposes a joint communication and sensing beamfo…
Cloud Native System for LLM Inference Serving
Minxian Xu, Junhan Liao, Jingfeng Wu +3
Large Language Models (LLMs) are revolutionizing numerous industries, but their substantial computational demands create challenges for efficient deployment, particularly in cloud…
Unlock the Potential of Fine-grained LLM Serving via Dynamic Module Scaling
Jingfeng Wu, Yiyuan He, Minxian Xu +3
The rise of large language models (LLMs) has created new opportunities across various fields but has also introduced significant challenges in resource management. Current LLM serv…
C-Koordinator: Interference-aware Management for Large-scale and Co-located Microservice Clusters
Shengye Song, Minxian Xu, Zuowei Zhang +5
Microservices transform traditional monolithic applications into lightweight, loosely coupled application components and have been widely adopted in many enterprises. Cloud platfor…
BrownoutServe: SLO-Aware Inference Serving under Bursty Workloads for MoE-based LLMs
Jianmin Hu, Minxian Xu, Kejiang Ye +1
In recent years, the Mixture-of-Experts (MoE) architecture has been widely applied to large language models (LLMs), providing a promising solution that activates only a subset of t…
BucketServe: Bucket-Based Dynamic Batching for Smart and Efficient LLM Inference Serving
Wanyi Zheng, Minxian Xu, Shengye Song +1
Large language models (LLMs) have become increasingly popular in various areas, traditional business gradually shifting from rule-based systems to LLM-based solutions. However, the…
Auto-scaling Approaches for Microservice Applications: A Survey and Taxonomy
Minxian Xu, Junhan Liao, Linfeng Wen +4
Microservice applications are created as loosely coupled application components and they leverage cloud elasticity to reduce costs and increase development speed. However, microser…
SealOS+: A Sealos-based Approach for Adaptive Resource Optimization Under Dynamic Workloads for Securities Trading System
Haojie Jia, Zhenhao Li, Gen Li +2
As securities trading systems transition to a microservices architecture, optimizing system performance presents challenges such as inefficient resource scheduling and high service…
Green Robotic Mixed Reality with Gaussian Splatting
Chenxuan Liu, He Li, Zongze Li +5
Realizing green communication in robotic mixed reality (RoboMR) systems presents a challenge, due to the necessity of uploading high-resolution images at high frequencies through w…
Clutter Resilient Occlusion Avoidance for Tightly-Coupled Motion-Assisted Detection
Zhixuan Xie, Jianjun Chen, Guoliang Li +4
Occlusion is a key factor leading to detection failures. This paper proposes a motion-assisted detection (MAD) method that actively plans an executable path, for the robot to obser…