Pujith Kachana
Papers
2
Total Citations
4
H-Index
2
About
Pujith Kachana is a researcher at the forefront of embodied AI and 3D scene understanding, with a focus on bridging the gap between natural language and robotic perception. His work centers on interactive referential grounding—enabling robots to interpret and act upon spatial language in complex, real-world 3D environments. In his highly cited paper, **"SORT3D: Spatial Object-centric Reasoning Toolbox for Zero-Shot 3D Grounding Using Large Language Models"** (2025, 2 citations), Kachana introduced a novel framework that leverages large language models to reason about spatial relations and object attributes without task-specific training, a critical step for robots operating alongside humans in cluttered scenes. Complementing this, his work **"IRef-VLA: A Benchmark for Interactive Referential Grounding with Imperfect Language in 3D Scenes"** (2025, 2 citations) addresses the practical challenge of noisy or ambiguous human instructions, providing a standardized benchmark for evaluating how well vision-language-action models can navigate and interact in indoor environments. Though early in his career, Kachana’s contributions are already shaping how robots understand and execute complex, free-form commands, laying essential groundwork for more intuitive human-robot collaboration.
Research Focus
Key Achievements
Top Papers
- 1
- 2