Hong-Wu Peng

Most-cited papers

1. Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

   Tian-Le Cai, Yu-Hong Li, Zheng-Yang Geng, Hong-Wu Peng +3 more

   2024International Conference on Machine Learning1,038 citations

   Semantic Scholar

   Save

2. Complete-muE: Optimal Hyperparameter Transfer and Scaling for MoE Models

   Hong-Wu Peng, Ohiremen L. Dibua, Yuan-Jun Xiong, Yi-Fan Gong +7 more

   2026arXiv.org2 citations

   Semantic Scholar

   Save

3. Multi-Head Low-Rank Attention

   Song-Tao Liu, Hong-Wu Peng, Zhiwei Zhang, Zheng-Yu Chen +1 more

   2026arXiv.org2 citations

   Semantic Scholar

   Save

4. Designing Draft Models for Speculative Decoding

   Reza Yazdani Aminabadi, Samyam Rajbhandari, Ahmad Awan, Cheng Li +34 more

   2 citations

   Semantic Scholar

   Save

5. Low-power enhanced system-on-chip design for sequential minimal optimisation learning core with tri-layer bus and butterfly-path accelerator

   Chih-Hsiang Peng, Po-Chuan Lin, S. Barma, Jhing-Fa Wang +3 more

   2015IET Computers & Digital Techniques1 citations Open access

   Semantic Scholar

   Save

6. Fast LLM Inference with Parallel Prompting

   Tianle Cai, Yu-Hong Li, Zheng-Yang Geng, Hong-Wu Peng +496 more

   0 citations

   Semantic Scholar

   Save