FW-Merging: Scaling Model Merging with Frank-Wolfe Optimization
Manuscript
Hao (Mark) Chen, Shell Xu Hu, Wayne Luk, Timothy Hospedales, Hongxiang Fan
[Paper]
Publications
You can also find the exhaustive list of my articles on my Google Scholar profile.
Advancing AI-assisted Hardware Design with Hierarchical Decentralized Training and Personalized Inference-Time Optimization
Manuscript
Hao (Mark) Chen, Zehuan Zhang, Wanru Zhao, Nicholas Lane, Wayne Luk, Hongxiang Fan
[Paper]
Progressive Mixed-Precision Decoding for Efficient LLM Inference
International Conference on Learning Representations (ICLR), 2025
Hao (Mark) Chen, Fuwen Tan, Alexandros Kouris, Royson Lee, Hongxiang Fan, Stylianos I. Venieris
[Paper] [Code]
Enhancing LLM-based Quantum Code Generation with Multi-Agent Optimization and Quantum Error Correction
Design Automation Conference (DAC), 2025
Charlie Campbell, Hao (Mark) Chen, Wayne Luk, Hongxiang Fan
[Paper]
Exploring Code Language Models for Automated HLS-based Hardware Generation: Benchmark, Infrastructure and Analysis
Asia and South Pacific Design Automation Conference (ASP-DAC), 2025
Jiahao Gai, Hao (Mark) Chen, Zhican Wang, Hongyu Zhou, Wanru Zhao, Nicholas Lane, Hongxiang Fan
[Paper]
Algorithm and Hardware Co-Design for Multi-Exit Dropout-based Bayesian Neural Networks
Manuscript
Hao (Mark) Chen, Liam Castelli, Martin Ferianc, Shuanglong Liu, Wayne Luk, Hongxiang Fan
[Paper]
Hardware-Aware Parallel Prompt Decoding for Memory-Efficient Acceleration of LLM Inference
Manuscript
Hao (Mark) Chen, Wayne Luk, Ka Fai Cedric Yiu, Rui Li, Konstantin Mishchenko, Stylianos I. Venieris, Hongxiang Fan
[Paper] [Code]
Hardware-Aware Neural Dropout Search for Reliable Uncertainty Prediction on FPGA
Design Automation Conference (DAC), 2024
Zehuan Zhang, Hongxiang Fan, Hao (Mark) Chen, Lukasz Dudziak and Wayne Luk
[Paper]
Web-based AI System for Medical Image Segmentation
Conference on Medical Image Understanding and Analysis (MIUA), 2023
Hao (Mark) Chen, Taowen Liu, Songyun Hu, Leyang Yu, Yiqi Li, Sihan Tao, Jacqueline Lee, Ahmed E. Fetit
[Paper]