TL;DR
Two systematic attempts to improve large pretrained models with minimal or zero modification to their weights via reinforcement learning on a frozen OpenVLA-7B using binary task-success rewards on LIBERO-Goal reveal a common ceiling.
LingBot-VA 2.0 is presented, a video-action foundation model built from the ground up for embodiment, which introduces a semantic visual-action tokenizer, which aligns visual representations with both semantics and actions, improving instruction following and action precision in subsequent policy learning.
A mechanism-by-mechanism transition to the recurrent behavior-cloning baseline shows that replacing the predictive pathway with direct observation input produces the largest single performance drop, accounting for approximately $70\% of the endpoint gap.
This work proposes Online Residual Policy Adaptation (ORPA), a framework that enables immediate, feedback-driven correction of robot actions without modifying the underlying policy parameters.
Hierarchical Robotic Control (HiRoC) is proposed, a hierarchical post-training framework that decouples high-level task planning from low-level action execution and aligns the executor with planner-generated subgoals before reinforcement learning, mitigating the distribution misalignment between planning and execution.
The results provide stronger evidence that RL-only bootstrapping can create the first usable language-conditioned controller for a genuinely novel embodiment, and use reinforcement learning in simulation with dense geometric rewards computed from simulator state.
Learning long-horizon robot manipulation remains difficult and time-consuming, especially under sparse rewards due to inefficient exploration and reward assignment. We present a minimal integration of large language models (LLMs) with reinforcement learning (RL) in which the LLM is used strictly as an online action pro...
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
