Haw-Shiuan Chang

Haw-Shiuan Chang is a Research Fellow in the Department of Computer Science at the University of Massachusetts Amherst, based in Room 366 of the Computer Science Building. His research focuses on advancing natural language processing and artificial intelligence through innovative modeling techniques.

His primary research interests include:

• Natural Language Processing (core focus)
• Machine Learning optimization
• Large Language Model efficiency
• Text generation factuality and diversity
• Computational creativity measurement

Recent publications (2021-2025) reveal a strong emphasis on resolving fundamental limitations in language models, particularly addressing entropy extrapolation, softmax bottlenecks, and multi-mode distribution representation. His work bridges theoretical modeling with practical applications in story generation, recommendation systems, and scientific document processing.

No scientific awards or mentoring activities are documented in the provided materials. Research is conducted within the Computer Science department infrastructure, with no specific lab or team affiliations mentioned.