REVERSAL-BENCH: A Reversibility Axis and Reset Oracle for Measuring the Reset-Free RL Cliff

Abstract:A central goal of autonomous reinforcement learning is continuous policy training without external resets. However, existing paradigms largely depend on underlying environmental reversibility, a property absent in real world manipulation, where events such as pushing objects off tables or spilling granular substances cannot be undone. We introduce REVERSAL-BENCH, a benchmark that controls reversibility via a continuous parameter $\rho \in [0, 1]$ and provides a reset oracle, a ground-truth verification mechanism to test state recoverability across eight manipulation settings in five physics engines. Evaluating a broad spectrum of policy architectures, including standard actor-critic algorithms, safe RL, and specialized reset-free frameworks, reveals a sharp reversibility cliff: reset-free agents are consistently absorbed into irrecoverable states as $\rho$ increases, whereas episodic agents maintain steady learning. We see this failure mode across autonomous reset-free baselines and constrained RL. Because reset-free agents lack external resets, any transition into an irrecoverable state results in permanent absorption, leaving the agent trapped where further learning halts. We show that this absorption phenomenon persists in full physics simulations under learned manipulation policies. By evaluating against geometrically identical reversible counterparts, we confirm that this breakdown is causally driven by irreversibility rather than obstacle complexity. We release the benchmark suite, a large multi-simulator dataset labeled with recoverability and a reset oracle. We also evaluate a safety shield that intervenes before irreversible failures occur, showing that while recoverability can be predicted accurately, active recovery primarily succeeds only when the agent can physically steer clear of the trap

Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)

arXiv:2609.17745 [cs.LG]

(or arXiv:2609.17745v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2609.17745 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.