My research focuses on making AI more efficient, interpretable, and accessible. Here are my publications and areas of interest.
Developing methods to reduce computational requirements of neural networks while maintaining performance. Focus on quantization, pruning, and knowledge distillation.
Creating AI systems that can understand and reason across different modalities including text, images, audio, and video.
Building AI systems that are transparent and explainable, enabling users to understand and trust model predictions.
Alex Chen, Sarah Zhang, Michael Roberts
We propose a novel sparse attention mechanism that reduces computational complexity while maintaining accuracy for vision transformers deployed on edge devices.
Alex Chen, Emily Wang
A new approach to multimodal learning that uses cross-attention mechanisms to effectively fuse information from text, images, and structured data.
Alex Chen, David Liu, Jennifer Kim
We extend LoRA to efficiently fine-tune models with 100B+ parameters using a hierarchical low-rank decomposition approach.
Alex Chen
A comprehensive survey of real-time object detection methods, covering architectures, optimization techniques, and deployment strategies.
Stanford AI Lab
Research Collaboration
Google DeepMind
Internship Project
NVIDIA Research
Open Source