Safety
The Robot Is Not Its Description: GaugeBench for Representation Robustness in Morphology-Aware Policies
arXiv:2610.07597v1 Announce Type: cross Abstract: A robot description does more than specify a physical mechanism: it also encodes arbitrary conventions, such as joint-axis direction, joint-angle zero
arXiv:2610.07597v1 Announce Type: cross Abstract: A robot description does more than specify a physical mechanism: it also encodes arbitrary conventions, such as joint-axis direction, joint-angle zero, and the order and names of links and joints. Morphology-aware policies consume interfaces built from these descriptions, yet cross-embodiment evaluation typically changes the robot while keeping those conventions fixed. This leaves a simple question unanswered: does behavior survive when the robot stays fixed but its description changes? GaugeBench isolates this case by rewriting a fixed mechanism under physically equivalent conventions, verifying that its physics and policy interface are preserved, and then evaluating the same policy weights. The result is stark: three MetaMorph policies score 4030.6 on 80 familiar robots, but only 51.6 when those same robots are equivalently re-described, while 98 genuinely held-out robots score 1489.6. A new description can therefore be more damaging than a new robot. Tracing the failure reveals that axis reversal alone reproduces the collapse, joint-angle zero changes are nearly harmless, and reordering lies between them; moreover, changing joint-state and torque coordinates alone is sufficient to cause the failure, while changing description-derived features alone is not. The same phenomenon appears in ModuMorph and an unrelated PyBullet framework. Yet it is not irreversible: exact two-description transport restores the original controller, and training across equivalent axis conventions raises retained return under axis reversal from 3.6% to 80.6%. Together, these results separate mechanism robustness from representation robustness and show that cross-embodiment evaluation should test both.
Related
- Robot-GST: geometry-aware spatial-temporal robot policy representation and evaluation
- CoWAM: Coordination Contracts for Selective Policy Intervention with WAMs
- Agile-WAM: An Agile Tactile World Action Model for Contact-Rich Robot Control
- When Does Test-Time Physical Diagnosis Pay? A Frozen Policy Buys Evidence It Never Reads
- A Mechanistic Analysis of Sim-and-Real Co-Training in Generative Robot Policies
Source: arXiv cs.LG | 2026-10-07