Penghui Yang
Scholar

Penghui Yang

Google Scholar ID: OuxqnTAAAAAJ
CCDS, Nanyang Technological University
Machine Learning
Citations & Impact
All-time
Citations
36
 
H-index
3
 
i10-index
1
 
Publications
8
 
Co-authors
2
list available
Resume (English only)
Academic Achievements
  • Publications: 1. AutoMAT: A Hierarchical Framework for Autonomous Alloy Discovery (arXiv 2025); 2. LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification (ES-FoMo ICML 2025); 3. Sailor2: Sailing in South-East Asia with Inclusive Multilingual LLMs (arXiv 2025); 4. Dual-Head Knowledge Distillation: Enhancing Logits Utilization with an Auxiliary Head (KDD 2025); 5. Multi-Label Knowledge Distillation (ICCV 2023)
Research Experience
  • Collaborated closely with Dr. Ming-Kun Xie and Prof. Lei Feng, and currently working closely with Dr. Cunxiao Du.
Education
  • PhD: College of Computing and Data Science, Nanyang Technological University, supervised by Prof. Bo An; B.Sc.: Computer Science, Nanjing University of Aeronautics and Astronautics, advised by Prof. Sheng-Jun Huang
Background
  • Research Interests: Developing efficient and scalable methods for accelerating and compressing machine learning models. Current focus is on speculative decoding to speed up LLM inference, while past work involved knowledge distillation for accelerating computer vision models. Aim to push the boundaries of fast and lightweight AI systems, making AI models more practical and widely accessible.
Miscellany
  • Personal Interests: Not provided