Crafter

The 'Crafter' dataset/benchmark contains high-dimensional observation spaces used to evaluate model-based reinforcement learning agents, particularly in their ability to learn abstract representations for effective planning and control.

Papers using Crafter (6)

Subgoal Graph-Augmented Planning for LLM-Guided Open-World Reinforcement Learning2025Autocontext: Instance-level Context Learning For LLM Agents2025Learning When to Plan: Efficiently Allocating Test-Time Compute for LLM Agents2025Goal-Guided Efficient Exploration via Large Language Model in Reinforcement Learning2025HERAKLES: Hierarchical Skill Compilation for Open-ended LLM Agents2025LLM Augmented Hierarchical Agents2023 · 1 cites