MAGMA-Bench: Evaluating Long-Horizon Embodied Agents under Dynamic Tasks and Execution Failures
From the abstract
Long-horizon embodied agents must maintain task-relevant information, coordinate heterogeneous robots, and adapt their decisions as execution unfolds. Existing benchmarks investigate dynamic task execution and embodied multi-agent coordination, but provide limited evaluation of their intersection.
From the abstract. Our summary is in progress.
The full text is not republished here because the paper's license does not allow it. Read the original on arXiv.