What Visual Representations Do World Action Models Need?
What Visual Representations Do World Action Models Need?
This research explores the visual representations required for World Action Models (WAMs), demonstrating that general-purpose representations provide broader robustness and outperform task-specific ones. The study evaluates five integration mechanisms and feature combinations, finding that channel-wise joint prediction achieves the highest success rate. The R$^2$-WAM model achieves 84.7% success on RoboCasa GR1 Tabletop and 92.8% on RoboTwin 2.0, improving over VAE-only baselines by 5.9 and 2.2 percentage points, respectively.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
Reported success rate
84.7%
View original evidence
achieves state-of-the-art performance on RoboCasa GR1 Tabletop with an 84.7% success rate
Open source S5
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
des a smaller additional gain than the initial improvement from adding a strong general-purpose representation to the VAE-only baseline. We instantiate the resulting Representation Recipe as R$^2$-WAM. It achieves state-of-the-art performance on RoboCasa GR1 Tabletop with an 84.7% success rate and reaches 92.8% on RoboTwin 2.0, improving over the VAE-only baselines by 5.9 and 2.2 percentage points, respectively. Take
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org