On Monday, January 12, Luciana Benotti will deliver the seminar in China: Stereotypes in ‘reasoning’ language models, focusing on how language models generate reasoning processes that can amplify stereotypes and unwarranted generalizations.
During the seminar, we will collect stereotypes live and test the EDIA framework in Chinese, exploring the challenges these systems face when operating across different linguistic and cultural contexts.
The activity takes place at the invitation of Fenrong Liu, distinguished professor at Tsinghua University and researcher, working on preference logic, multi-agent systems, logical reasoning for large language models (LLMs), neuro-symbolic AI, and the history of logic in China. Fenrong Liu will also be organizing an upcoming workshop at ICLR 2026, the leading conference in machine learning, which will be held for the first time in Latin America, in Rio de Janeiro, Brazil. For more information of the whorkshop: https://sites.google.com/view/iclr-2026-llmreasoning/
