Sriram Yenamandra
Papers
4
Total Citations
103
H-Index
4
About
Sriram Yenamandra is a leading researcher in embodied AI, whose work bridges the gap between language, vision, and physical robot action. His primary research areas include embodied question answering, open-vocabulary mobile manipulation, and lifelong navigation. Yenamandra’s major contributions center on creating systems that allow robots to understand and interact with complex, real-world environments using natural language and visual cues. He led the development of **OpenEQA** (46 citations), a modern benchmark for embodied question answering that tests an agent’s ability to reason about its surroundings, and **GOAT: GO to Any Thing** (37 citations), a universal navigation system enabling mobile robots to find any object in unseen spaces. He is also the architect of **HomeRobot** (13 citations), an open-source platform for open-vocabulary mobile manipulation, and **GOAT-Bench** (7 citations), a benchmark for multi-modal lifelong navigation. His work is notable for its focus on practical, deployable systems that move beyond single-task, single-modality approaches, setting new standards for how robots can operate autonomously in homes and warehouses.
Research Focus
Key Achievements
Top Papers
- 1OpenEQA: Embodied Question Answering in the Era of Foundation Models46 citations · 2024
- 2GOAT: GO to Any Thing37 citations · 2024
- 3HomeRobot: Open-Vocabulary Mobile Manipulation13 citations · 2023
- 4GOAT-Bench: A Benchmark for Multi-Modal Lifelong Navigation7 citations · 2024