Herbert Wertheim College of Engineering
Sun Lab
Led by Jingwei Sun · Computer and Information Science and Engineering
What we study
The Sun Lab studies how to make AI systems work better on devices with limited power, like phones or sensors. They focus on making AI faster, more reliable, and personalized for users.
What working here is like
- Computational
- No organisms (data only)
- Coding required
- Papers
- 5
Recent papers, explained
- EC-SpMV: Toward Efficient SpMV in Sparse LLMs via Block Extraction and Compressed Storage
2026 · ACM Transactions on Architecture and Code Optimization
Speeds up sparse matrix operations in large language models by reorganizing data storage.
- Flash-DWC: Making Depthwise Convolution Compute-Efficient on GPUs
2026 · ACM Transactions on Architecture and Code Optimization
Improves the speed of a key neural network operation on GPUs by optimizing data handling.
Show 3 more papersShow fewer
- GNNPilot: A Holistic Framework for High-Performance Graph Neural Network Computations on GPUs
2025 · ACM Transactions on Architecture and Code Optimization
Optimizes graph neural network performance on GPUs by improving data organization and parallel processing.
- Monolithic 3-D Integration of 2T0C DRAM and 1T1R RRAM for Accelerating Dynamic/Static Matrix Computation in Transformer Network
2025 · IEEE Transactions on Electron Devices
Combines two types of memory in a 3D chip to speed up calculations in transformer networks.
- HFProxy: Synthesizing Portable and High-Fidelity Proxy Applications for MPI Programs
2025 · ACM Transactions on Architecture and Code Optimization
Creates simplified versions of complex programs to test performance without full-scale testing.
Plain-language summaries are AI-written and reviewed by us; check the paper for details.
How to reach out
Jingwei Sun hasn't set up this page yet. Their UF Scholars profile lists their publications.