Yanxuan Yu

Efficient Inference for Language Models

Query-Aware Caching

Sparse Attention

Systems for Large-Scale NLP

Real-Time Inference

CUDA Kernel Optimization

Multimodal Biomedical Applications

Signal Processing for Neural Data

Machine Learning for Biomedical Engineering

Embedded AI Systems

Small Language Models

TinyLLaMA

Lightweight Transformer Runtime

Token Selection

Page-Level Cache Scheduling

Memory-Constrained Serving