Ellie Pavlick
Corvallis Environmental Center, Brown University, John Brown University
Papers
7
Total Citations
105
H-Index
6
About
Ellie Pavlick’s research lies at the intersection of natural language processing and robotics, focusing on how robots can understand and act upon complex, human-like language in real-world environments. Her major contributions center on grounding natural language to spatial and temporal tasks, enabling robots to interpret commands that reference never-before-seen landmarks, sequential constraints, and abstract spatial relations. For instance, her work on “Robot Object Retrieval with Contextual Natural Language Queries” (43 citations) advances robots’ ability to locate objects using contextual cues, while her studies on grounding language to landmarks in arbitrary outdoor environments (17 citations) and non-Markovian tasks (19 citations) push beyond simple goal-oriented commands to handle temporal specifications like “go around the lake and then travel north.” Pavlick also explores affordance-based retrieval and hierarchical planning with state abstractions, addressing challenges in partially observed, city-scale settings. Her research is notable for reducing the barrier between humans and robots, making robotic interaction more intuitive and adaptable. With over 100 citations across her most-cited works, Pavlick’s contributions are shaping how robots navigate and assist in human-centric spaces, from indoor retrieval to outdoor navigation.
Research Focus
Key Achievements
Top Papers
- 1Robot Object Retrieval with Contextual Natural Language Queries43 citations · 2020
- 2
- 3Grounding Language to Landmarks in Arbitrary Outdoor Environments17 citations · 2020
- 4
- 5Affordance-based robot object retrieval7 citations · 2021
- 6
- 7Planning with State Abstractions for Non-Markovian Task Specifications2 citations · 2019