Spandana Gella

Papers

Total Citations

H-Index

About

Spandana Gella is a leading researcher at the intersection of natural language processing and embodied AI, with a primary focus on building interactive agents that can communicate and collaborate with humans in physical environments. Her most impactful contribution is the creation of **TEACh (Task-driven Embodied Agents that Chat)**, a benchmark dataset comprising over 3,000 human-human interactive dialogues. This resource is foundational for studying how robots can not only follow instructions but also use conversation to resolve ambiguity, ask clarifying questions, and recover from mistakes during real-world tasks. The 2022 iteration of this work has garnered **89 citations**, underscoring its influence in the field. Gella’s research addresses a critical gap in robotics: enabling natural, bidirectional communication between humans and machines. By advancing the study of situated dialogue, her work paves the way for more capable and trustworthy embodied agents that can operate seamlessly in human spaces, making her a key figure in the development of next-generation human-robot interaction systems.