Hanchi Sun is a first-year PhD student in the Computer Science and Engineering department at Lehigh University. His advisor is Professor Lichao Sun. Before that, he was a master’s and undergraduate student at Lehigh University studying Computer Science and Mathematics.

His research interests center around large language models, with a focus on efficient training and inference.

Research Interests

Large Language Models, Efficient Training and Inference, Natural Language Processing

News

• [May-August 2026] I will join Adobe as a research intern.

• [April 2026] MegaTrain: Full Precision Training of 100B+ Parameter Large Language Models on a Single GPU is available on arXiv.

• [March 2026] Expert Threshold Routing for Autoregressive Language Modeling with Dynamic Computation Allocation and Load Balancing is available on arXiv.

• [May-August 2025] I joined Adobe as a research intern.

• [September 2024] SpecHub: Provable Acceleration to Multi-Draft Speculative Decoding has been accepted to EMNLP 2024 (main).

• [September 2024] Virtual Context Enhancing Jailbreak Attacks with Special Token Injection has been accepted to Findings of EMNLP 2024.