Sue Felshin
Papers
1
Total Citations
5
H-Index
1
About
Sue Felshin is a research scientist whose work lies at the intersection of natural language processing, robotics, and human-robot interaction, with a particular focus on enabling machines to understand and act upon grounded, context-rich language. Her key contributions center on developing frameworks that allow robots to interpret natural language instructions by integrating visual perception with accrued linguistic context from past interactions. In her most cited work, "Temporal Grounding Graphs for Language Understanding with Accrued Visual-Linguistic Context" (2017), Felshin introduced a novel approach that enables robots to ground ambiguous or temporally dependent commands by building a structured representation of both current visual data and historical dialogue. This work, which has garnered 5 citations, addresses a fundamental challenge in robotics: bridging the gap between human language and a robot's situated knowledge. Felshin's research is notable for its practical emphasis on creating systems that learn from ongoing interactions, moving beyond one-shot instruction following toward more adaptive, conversational AI. Her contributions are particularly valuable for students and researchers working on embodied agents, situated dialogue, and the integration of perception with language understanding in real-world environments.
Research Focus
Key Achievements
Top Papers
- 1