AI2-THOR

AI-2-THOR is a dataset and benchmark that contains a variety of simulated indoor environments used to evaluate the performance of AI systems in understanding and interacting with complex tasks involving multiple objects and actions.

Papers using AI2-THOR (3)

LaMMA-P: Generalizable Multi-Agent Long-Horizon Task Allocation and Planning with LM-Driven PDDL Planner2024 · 1 citesLLaMAR: Long-Horizon Planning for Multi-Agent Robots in Partially Observable Environments2024 · 1 citesLoTa-Bench: Benchmarking Language-oriented Task Planners for Embodied Agents2024