Interactive Planning

Interactive planning focuses on developing systems capable of generating and adapting plans in dynamic environments, often involving human-computer interaction or multi-agent collaboration. Current research emphasizes integrating reinforcement learning, particularly hierarchical and multi-agent approaches, with large language models and other advanced search algorithms like large neighborhood search to improve efficiency and robustness in complex scenarios such as robotics, dialogue systems, and education. This field is significant for its potential to create more adaptable and intelligent systems across diverse applications, ranging from warehouse automation to personalized tutoring and open-world robotic task completion.