M.H Shridhar

Papers

Total Citations

H-Index

About

M.H. Shridhar is a leading researcher at the intersection of robotics, computer vision, and natural language processing, with a focus on enabling robots to understand and act upon human instructions. His work is particularly notable for pioneering the application of Transformer architectures to robotic manipulation, a domain traditionally constrained by limited and expensive data. In his highly influential 2022 paper, "Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation" (48 citations), Shridhar demonstrated that with the right problem formulation, Transformers can indeed scale effectively for manipulation tasks, offering a new paradigm for multi-task learning in robotics. Earlier, his 2018 work on "Interactive Visual Grounding of Referring Expressions for Human-Robot Interaction" (6 citations) introduced INGRESS, a system that enables robots to follow natural language instructions to pick and place objects by grounding referring expressions in visual data. This foundational contribution to human-robot interaction showcases Shridhar’s ability to bridge language understanding and physical action. His research is widely cited for its practical impact, advancing the goal of creating robots that can seamlessly collaborate with humans in real-world environments.