Jessica Borja-Diaz

Papers

Total Citations

H-Index

About

Jessica Borja-Diaz is an emerging robotics researcher whose work sits at the compelling intersection of language grounding, visual affordance learning, and sample-efficient robot skill acquisition. Her research addresses one of the field's most pressing challenges: enabling robots to operate intelligently in unstructured, human-centered environments without requiring prohibitive amounts of labeled data or constant human intervention. Her most impactful contribution, "Grounding Language with Visual Affordances over Unstructured Data," has accumulated 66 citations since 2023, demonstrating significant community interest in her approach to connecting Large Language Models with robot manipulation through visual affordances. This work tackles the practical limitations of deploying language-conditioned robotic systems at scale. Complementing this, her 2022 paper "Affordance Learning from Play for Sample-Efficient Policy Learning" (29 citations) introduces a self-supervised framework that allows robots to extract meaningful object interaction knowledge from unstructured play data — dramatically reducing the data burden typically required for multi-task learning. Together, these contributions reflect Borja-Diaz's broader mission to make robots more autonomously capable and data-efficient, positioning her as a promising voice in the growing field of language-guided robot learning and embodied AI.