FedEL: Federated Elastic Learning for Heterogeneous Devices
M.Phil. Student at HKU
I study the mathematical foundations of large language models and agentic systems, with a focus on how attention, depth, and scale shape reasoning and emergent capabilities.
I am a 1st-year M.Phil. student at University of Hong Kong , where I am fortunate to be advised by Prof. Yingyu Liang and Prof. Andrew F. Luo.
I am also a Research Engineer Intern at Moonshot AI working on LLM Infra.
Research
Selected Publications
* denotes equal contribution or alphabetical order.
Circuit Complexity Bounds for RoPE-based Transformer Architecture*
Hsr-enhanced Sparse Attention Acceleration*
Bypassing the Exponential Dependency: Looped Transformers Efficiently Learn In-context by Multi-step Gradient Descent*