Yury Sulsky

Papers

Total Citations

H-Index

About

Yury Sulsky is a leading researcher in artificial intelligence, specializing in generalist agents, multi-modal learning, and large-scale reinforcement learning. His most notable contribution is the development of **Gato**, a groundbreaking generalist agent introduced in the 2022 paper *"A Generalist Agent"* (66 citations). Inspired by advances in large language models, Sulsky and his team created a single neural network capable of performing hundreds of tasks across diverse modalities—including playing Atari games, captioning images, controlling a robotic arm, and stacking blocks—all with the same set of weights. This work demonstrated that a unified policy could effectively handle multiple embodiments and reward functions, challenging the traditional paradigm of task-specific agents. Gato’s impact is profound, offering a scalable path toward artificial general intelligence by treating diverse tasks as sequences of tokens. Sulsky’s research has garnered significant attention for bridging the gap between language models and embodied AI, making him a key figure in the push for versatile, multi-task agents. His work continues to inspire new directions in foundation models for robotics and interactive environments.