Unexpected Behavior
Unexpected behavior, encompassing deviations from expected patterns in various systems, is a growing area of research focusing on identifying, quantifying, and mitigating its impact. Current studies explore this across diverse domains, including the randomness of large language models, the robustness of deep learning algorithms under noise and privacy constraints, and the responses of children to unexpected robot actions. This research aims to improve model reliability, enhance the understanding of human-computer interaction, and develop more robust and adaptable systems in various applications, from recommendation systems to autonomous driving.
Papers
May 31, 2024
May 14, 2024
October 19, 2023
May 12, 2023
June 16, 2022