Publications
You can also find my articles on my Google Scholar profile.
Journal Articles
Efficient Model-Based Reinforcement Learning for Robot Control via Online Learning
An online model-based reinforcement learning algorithm for efficient real-world robot control using dynamics models learned from interaction data.
Conference Papers
SALAAD: Sparse And Low-Rank Adaptation via ADMM for Large Language Model Inference
A plug-and-play sparse and low-rank adaptation framework for large language model inference under heterogeneous memory budgets.
Grounding Generative Policies in Physics: Optimization-Guided Diffusion for Robot Control
Inference-time optimization grounds diffusion policies in robot reachability, collision-avoidance, and controller-level feasibility constraints without retraining.
Preprints
Learning Structured Sparsity for Large Models: Bridging Model Compressibility and Efficient Inference
S-SALAAD learns structured sparse and low-rank weights to enable elastic model compression and efficient CPU and GPU inference across memory budgets.
Rethinking Routing Exploration in Modern Mixture-of-Experts
MirrorMoE adapts the routing exploration space to help shared-expert models realize the benefits of fine-grained expert segmentation.