Winston Zeng

Papers on Lattice

Total citations

Topics

h-index

Publication activitypapers/week, last 8 weeks

Research focus

Interpretability & Mechanistic Interp (1)Natural Language Processing (1)Open-Source Models & Weights (1)

Frequent co-authors

Ali Emami (1)Jinho Choi (1)

Papers (1)

Jul 14, 2026

Winston Zeng +21w ago

What Models Express, Suppress, and Resist: Auditing Open-Weight LLMs with Persona Vectors

Persona vectors expose the hidden complexities of LLM behavior, revealing that steering can unlock traits like hyperbole and hallucination that models typically suppress.

Winston Zeng, Ali Emami, Jinho Choi

Interpretability & Mechanistic Interp Natural Language Processing Open-Source Models & Weights

Search

Winston Zeng

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (1)