Christina Sarkisyan
Papers
1
Total Citations
8
H-Index
1
About
Christina Sarkisyan is a researcher at the forefront of embodied artificial intelligence, exploring how large language models (LLMs) can bridge the gap between abstract reasoning and physical action. Her work centers on the critical evaluation of pretrained LLMs in embodied planning tasks—investigating whether these models can effectively translate language understanding into real-world, sequential decision-making for robots and agents. Sarkisyan’s most-cited paper, "Evaluation of Pretrained Large Language Models in Embodied Planning Tasks" (2023), has garnered 8 citations and serves as a foundational benchmark for the field. In this study, she systematically tested models like GPT-3 on complex, multi-step tasks, revealing both their surprising zero-shot capabilities and their limitations in long-horizon planning. This work has informed subsequent research on grounding language in physical environments, influencing the design of more robust AI systems. By rigorously assessing model performance, Sarkisyan has helped shape the trajectory of embodied AI, offering clear metrics and insights that guide both academic inquiry and practical deployment. Her contributions are essential reading for anyone interested in the intersection of natural language processing and robotics.
Research Focus
Key Achievements
Top Papers
- 1