Office World

The 'Office World' dataset/benchmark contains a simulated environment used to evaluate the effectiveness of LLM-assisted semantic option discovery in Deep Reinforcement Learning.

Papers using Office World (3)

Sample-efficient Neuro-symbolic Proximal Policy Optimization2026LLM-assisted Semantic Option Discovery for Facilitating Adaptive Deep Reinforcement Learning2026ComputerRL: Scaling End-to-End Online Reinforcement Learning for Computer Use Agents2025