Karthik Mahadevan
Papers
5
Total Citations
72
H-Index
4
About
Karthik Mahadevan is a leading researcher in human-robot interaction (HRI), focusing on how robots can perceive, interpret, and respond to subtle human cues for more natural collaboration. His work bridges expressive robot behavior, intuitive teleoperation, and interactive learning. Mahadevan’s most cited paper, “Generative Expressive Robot Behaviors using Large Language Models” (2024, 49 citations), pioneers the use of LLMs to generate nuanced, context-aware robot gestures—like nodding or excusing oneself—enabling robots to communicate more effectively in social spaces. In “Stargazer” (2023, 10 citations), he developed an interactive camera robot that reads subtle instructor cues to autonomously capture how-to videos, reducing the need for a human cameraperson. His earlier work, “Mimic” (2022, 6 citations), introduced in-situ recording and reuse of demonstrations to simplify remote robot teleoperation, addressing key usability challenges. Mahadevan’s research consistently emphasizes user agency and safety, as seen in his “Fight-or-Flight” (2018) study exploring HRI safety mechanisms through a Research through Design approach. His recent “ImageInThat” (2025) work further advances human-robot instruction by leveraging foundation models for image-based communication. With a growing citation impact and contributions that directly improve how humans teach and interact with robots, Mahadevan is shaping the future of intuitive, expressive, and safe robotic systems.
Research Focus
Key Achievements
Top Papers
- 1Generative Expressive Robot Behaviors using Large Language Models49 citations · 2024
- 2
- 3
- 4ImageInThat: Manipulating Images to Convey User Instructions to Robots4 citations · 2025
- 5"Fight-or-Flight"3 citations · 2018