Qihua Zhou

Latest

D^2MoE: Dual Routing and Dynamic Scheduling for Efficient On-Device MoE-based LLM Serving
DeNC: Unleash Neural Codecs in Video Streaming with Diffusion Enhancement
Mjölnir: Breaking the Shield of Perturbation-Protected Gradients via Adaptive Diffusion
FreePIH: Training-Free Painterly Image Harmonization with Diffusion Model
Collaborative Neural Architecture Search for Personalized Federated Learning
Model Decomposition and Reassembly for Purified Knowledge Transfer in Personalized Federated Learning
ParsNets: A Parsimonious Composition of Orthogonal and Low-Rank Linear Networks for Zero-Shot Learning
On the Robustness of Neural-enhanced Video Streaming Against Adversarial Attacks
Chiron: A Robustness-aware Incentive Scheme for Edge Learning via Hierarchical Reinforcement Learning
PASS: Patch Automatic Skip Scheme for Efficient On-device Video Perception
Development of Deep Learning Algorithms for Automated Scoliosis and Abnormal Posture Screening Using 2D Back Image
Graph Knows Unknowns: Reformulate Zero-shot Learning as Sample-level Graph Recognition
PASS: Patch Automatic Skip Scheme for Efficient Real-time Video Perception on Edge Devices
Tree Learning: Towards Promoting Coordination in Scalable Multi-Client Training Acceleration
Hierarchical Channel-spatial Encoding for Communication-efficient Collaborative Learning
Octo: INT8 Training with Loss-aware Compensation and Backward Quantization for Tiny On-device Learning
Petrel: Heterogeneity- aware Distributed Deep Learning via Hybrid Synchronization
Canary: Decentralized Distributed Deep Learning via Gradient Sketch and Partition in Multi-interface Networks
Falcon: Addressing Stragglers in Heterogeneous Parameter Server via Multiple Parallelism
Dual-view Attention Networks for Single Image Super-Resolution
Falcon: Towards computation-parallel deep learning in heterogeneous parameter server
Fast coflow scheduling via traffic compression and stage pipelining in datacenter networks
Swallow: Joint Online Scheduling and Coflow Compression in Datacenter Networks