Researchers at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed "Masked Inverse Reinforcement Learning (Masked IRL)," a machine learning approach that uses dual large language models (LLMs) to clarify ambiguous human instructions and filter out irrelevant environmental data, enabling robots to safely execute complex tasks. One LLM expands upon vague user prompts based on physical demonstration data, while a second LLM "masks" irrelevant environmental details by assigning them a score of "0" and prioritizing critical elements as "1" for the final algorithmic motion plan. The system reduces the required physical demonstration data by nearly five times and improves the identification of unstated user preferences by up to 15 percent. The research is slated for presentation at the June 2026 IEEE International Conference on Robotics and Automation. Future applications include integrating camera-based visual data for deployment in domestic, office, and industrial manufacturing settings.