I am Zeyu Huang, a PhD student (2024.01 - ) at the University of Edinburgh. I am very lucky to be supervised by Dr.Ivan Titov and Dr.Edoardo M. Ponti.

Before that, I obtained my master’s degree at the Ministry of Education (MOE) Engineering Research Center of Advanced Computer Application Technology, Beihang University, and my engineer degree (diplôme d’ingénieur) at École Centrale de Pékin, Groupe des Écoles Centrale (France), supervised by Zhang Xiong and Wenge Rong. I obtained my BSc degree in Information and Computing Science from Beihang University and Centrale Pékin.

I have previously worked on Model editing and Mixture-of-Experts. I am now broadly interested in AGI and Large Language Models (language is so important because it’s the interface between Humans and AI). My current goal is to become a full-stack LLM researcher. If you are interested in my research, I would be pleased to chat with you! You can directly send the email to zeroy.huang AT gmail DOT com.

Hight Lights

• Our paper, Stacking Your Transformers: A Closer Look at Model Growth for Efficient LLM Pre-Training, has been accepted by NeurIPS 2024 (spotlight)!

• Our paper, Unlocking Emergent Modularity in Large Language Models, gets the NAACL 2024 Outstanding Paper award!

• Happy to start my PhD at UoE!