M. Sharma
Papers
2
Total Citations
8
H-Index
2
About
M. Sharma is a rising force at the intersection of embodied AI and cognitive reasoning, pushing the boundaries of how machines understand and interact with the human world. Their work is defined by two major thrusts: creating large-scale, human-centered benchmarks for robotics and probing the spatial reasoning limits of large language models. Sharma’s flagship contribution, the BEHAVIOR-1K benchmark (2024, 6 citations), is a landmark achievement—a simulation environment grounded in a survey of 1,000 everyday activities people actually want robots to perform. This resource provides an unprecedented, realistic testbed for training and evaluating generalist robots, directly shaping the future of assistive technology. Complementing this, Sharma’s exploration of spatial reasoning in LLMs (2023, 2 citations) tackles a critical gap, investigating how these models handle numerical trajectory data—a skill essential for navigation and planning. Though early in their career, Sharma’s work demonstrates a rare ability to identify foundational challenges and build the tools to solve them, making them a researcher to watch in the drive toward truly capable, human-aware AI systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2