MAT-THOR

MAT-THOR is a benchmark dataset that contains multi-robot task planning scenarios used to evaluate the performance of planning systems in interpreting natural-language instructions and generating executable actions.

Papers using MAT-THOR (2)

Hierarchical Llm-based Multi-agent Framework With Prompt Optimization For Multi-robot Task Planning2026KGLAMP: Knowledge Graph-guided Language model for Adaptive Multi-robot Planning and Replanning2026