Feasible Task

"Feasible task" research focuses on ensuring that tasks assigned to artificial intelligence systems, particularly large language models and robots, are achievable given the system's capabilities and environmental constraints. Current research emphasizes developing methods to identify and reject infeasible tasks, improve the efficiency of feasible task execution (e.g., through optimized sampling strategies in MRI or efficient replanning algorithms), and generate feasible solutions in creative domains like engineering design. This work is crucial for enhancing the reliability and trustworthiness of AI systems across diverse applications, from medical imaging and robotics to engineering design and natural language processing.