Learning task progress from robot demonstrations provides a practical way to construct dense shaping rewards for robotic reinforcement learning. A common approach uses normalized demonstration time as progress labels, implicitly assuming monotonically increasing task progress. During policy exploration, however, actual...
Jun Nie, Zi-Tian Fan, Jia-Chen Zhang et al.· 0 citations
This work proposes MORPH: Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization, a staged framework with large-scale raw-sequence pretraining, Metric-Ordered Sequence (MOS) training, and final HPPO alignment.
Results show that a fixed-weight, self-evolving harness can revise, recover, and accumulate verified approaches while producing structured trajectories for future supervised and reinforcement learning.
Boxiu Li, Zi-Mo Wen, Yi-Jia Fan et al.· 2 citations· ⚡1
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.