Openpi Comet: Competition Solution For 2025 BEHAVIOR Challenge

TL;DR

Openpi Comet presents a competitive solution for the 2025 BEHAVIOR Challenge, demonstrating significant performance improvements through systematic training techniques and data analysis in long-horizon household tasks.

Contribution

The paper introduces a novel approach that leverages ablation studies and scaling strategies to enhance foundation models for complex embodied AI tasks.

Findings

Achieved 2nd place in the 2025 BEHAVIOR Challenge.

Validated the effectiveness of scaling in pre-training and post-training phases.

Provided practical insights and design recommendations for embodied AI applications.

Abstract

The 2025 BEHAVIOR Challenge is designed to rigorously track progress toward solving long-horizon tasks by physical agents in simulated environments. BEHAVIOR-1K focuses on everyday household tasks that people most want robots to assist with and these tasks introduce long-horizon mobile manipulation challenges in realistic settings, bridging the gap between current research and real-world, human-centric applications. This report presents our solution to the 2025 BEHAVIOR Challenge in a very close 2nd place and substantially outperforms the rest of the submissions. Building on $\pi_{0.5}$, we focus on systematically building our solution by studying the effects of training techniques and data. Through careful ablation studies, we reveal the scaling benefits in both the pre-training and post-training phases, leading to a validation Q-score of 0.345, significantly surpassing previous state-of-the-art performance. We summarize our practical lessons and design recommendations that we hope will provide actionable insights for the broader embodied AI community when adapting powerful foundation models to complex embodied scenarios. Project page: https://github.com/mli0603/openpi-comet