CLeAR had three papers accepted to IROS this year! Come join us in Detroit! Congrats to Tasbolat, Bowen, and Kaiqi. Find out more about the papers below:
- Large Language Models as Zero-Shot Human Models for Human-Robot Interaction: This research delves into the capabilities of large language models (LLMs), which have ingested immense amounts of text created by humans, to serve as instant human models in Human-Robot Interactions without prior training on specific tasks.
- Latent Emission-Augmented Perspective-Taking (LEAPT) for Human-Robot Interaction: This study addresses the concept of perspective taking, which involves understanding and assuming the viewpoint of another entity, be it visually or mentally.
- Refining 6-DoF Grasps with Context-Specific Classifiers: Here, we introduce GraspFlow, a method to enhance the precision of context-aware grasps. The challenge of formulating a grasp is seen as a sampling dilemma: the goal is to select from a probability distribution of successful grasps, influenced by the surrounding context.
