Papers

10

Total Citations

255

H-Index

6

About

Aniruddha Kembhavi is a leading researcher at the intersection of computer vision, natural language processing, and embodied AI, where he drives the development of generalist agents that can perceive, reason, and act in the physical world. As a key figure behind the Allen Institute for AI (AI2), his most impactful contribution is **Unified-IO 2**, the first autoregressive multimodal model capable of understanding and generating text, images, audio, and action—a foundational step toward truly unified AI systems. His work on **Open X-Embodiment** (119 citations) has been instrumental in consolidating robotic learning across diverse datasets, enabling large-scale, general-purpose robot policies. Kembhavi also created **AllenAct**, a widely adopted framework for embodied AI research (44 citations), and **ManipulaTHOR**, which advanced visual object manipulation in simulation. More recently, his **Phone2Proc** method bridges the sim-to-real gap by using a 10-minute phone scan to generate realistic training environments, while **Promptable Behaviors** personalizes robot policies to human preferences. With over 250 total citations and a consistent focus on scaling multimodal, interactive agents, Kembhavi’s work is shaping the future of robots that can navigate, manipulate, and collaborate in our chaotic, real-world environments.

Research Focus

Key Achievements

6
H-Index
10
Papers
255
Total Citations
26
Avg Citations/Paper
🏆 Most Cited Paper
Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration<sup>0</sup>
119 citations · 2024
📈 Most Prolific Year: 2024 (5 Papers)
🤝 Key Collaborators: 129
🏛 Institutions: Allen Institute, Seattle University, University of Washington, Allen Institute for Artificial Intelligence

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
    Harmonic Mobile Manipulation
    3 citations · 2024
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago