About Me

Hello, I'm Sonny George. I've just finishing my Master's at Brandeis University and am seeking research and other opportunities in the space of action-taking AI or software development generally.

Much of future impactful AI will require some form of active embodiment, reflection, and continual learning. In contrast, the “intelligence” of today's AI systems tends to be static and task-specific.

I am especially excited by the development of AI systems that:

Plan and act in uncertain environments
Adapt to novel tasks/situations with inference-time-feasible efficiency and without catastrophic forgetting
Are sufficiently interpretable and manipulable to be aligned with human preferences

Publications

Benchmarking Arbitrary Natural Language Tasks in 3D Open Worlds

Sonny George, Chris Sypherd, Rocco Ahching, and Dylan Cashman
Proceedings of the 6th Annual Embodied Artificial Intelligence Workshop at CVPR 2025 (Oral Spotlight)

Probing the Capacity of Language Model Agents to Operationalize Disparate Experiential Context Despite Distraction

Sonny George, Chris Sypherd, and Dylan Cashman
Findings of the 2024 Conference on Empirical Methods in Natural Language Processing

Incorporating Token Usage into Prompting Strategy Evaluation

Chris Sypherd, Sergei Petrov, Sonny George, and Vaishak Belle
ArXiv Preprint, May 2025

Presentations

sr-OLTHAD+SemanticSteve: Toward Any-Horizon Open-Ended Autonomy (Master's Capstone Presentation)

May 2025

Blog Posts

January 10, 2025: What is the proper level of I/O abstraction in world models (that generate simulated rollouts)?

A Few (humble) Hardware Projects

BC+PPO Learning for Simple Object Picking

Sonny George