$\chi_{0}$: Resource-Aware Robust Manipulation via Taming Distributional Inconsistencies

ArXi:2602.09021v2 Announce Type: replace-cross High-reliability long-horizon robotic manipulation has traditionally relied on large-scale data and compute to understand complex real-world dynamics. However, we identify that the primary bottleneck to real-world robustness is not resource scale alone, but the distributional shift among the human nstration distribution, the inductive bias learned by the policy, and the test-time execution distribution -- a systematic inconsistency that causes compounding errors in multi-stage tasks.