RM‑VLM introduces a method to generate reward machines (RMs) from natural‑language task descriptions using a large language model, and employs a vision‑language model as the labeling function to ground RM propositions over image observations. This approach eliminates the need for manually engineered labeling functions, enabling RM‑guided policy learning across diverse domains. Experiments on grid‑world navigation, robotic manipulation, and 3D object collection demonstrate that RM‑VLM consistently outperforms existing VLM‑based reward methods in task success rates.
arXiv:2606. 00083v1 Announce Type: cross Abstract: Reinforcement learning relies on accurate reward functions, which are often hand-crafted or even unavailable in real-world applications, such as robotics.
arXiv:2602. 19313v2 Announce Type: replace-cross Abstract: General-purpose robot learning requires dense, instruction-conditioned feedback that can distinguish meaningful task progress from stalled, failed, or partially completed behavior.
The paper introduces D&A (Describe & Assess), a prompt‑based method that separates vision‑language model progress estimation into a textual description of scene dynamics and a rubric‑based assessment. By requiring the model to explicitly define absolute success before scoring, D&A reduces confirmation bias and provides fine‑grained, partial progress metrics that accurately penalize failures and interpolate successes. The approach generalizes across tasks, embodiments, and benchmarks, matching or surpassing post‑trained reward models while enabling efficient downstream reinforcement learning.
The paper introduces D&A (Describe & Assess), a prompt-based method that separates progress estimation into a textual description of scene dynamics and a subsequent assessment using explicit rubrics. By requiring Vision‑Language Models to generate an explicit definition of absolute success before scoring, D&A anchors evaluations to an external criterion, reducing confirmation bias and providing fine‑grained, partial progress metrics. Empirical results show that D&A matches or surpasses post‑trained reward models across tasks, embodiments, and benchmarks, and its reward estimates support efficient downstream reinforcement learning.
arXiv:2604.00055v2 Announce Type: replace-cross Abstract: Existing robotic foundation policies are trained primarily via large-scale imitation learning. While such models demonstrate strong capabilit...
The paper introduces Stepwise Marginal Information Gain (MIG), an intrinsic process reward that evaluates how each reasoning step of a large language model (LLM) or vision-language model (VLM) improves the likelihood of the reference answer. MIG rewards only new likelihood maxima, preventing duplicate credit, and is combined with outcome, format, and self‑distillation objectives to guide training. Experiments on eight benchmarks show that this method outperforms outcome‑only reinforcement learning and improves accuracy by up to 4.8 points over binary‑reward training, including a 12.6‑point gain on MathVerse and a 12.9‑point advantage on vision‑language transfer at 7B parameters.
arXiv:2603.16065v3 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) has shown strong potential for improving robotic manipulation policies, yet its practical use remains bottlenecke...
arXiv:2608. 03875v1 Announce Type: cross Abstract: Designing effective reward functions remains a major bottleneck in Reinforcement Learning (RL).
arXiv:2608. 05954v1 Announce Type: new Abstract: Reinforcement Learning (RL) is a powerful but far from easy-to-use technique for policy learning.
arXiv:2502. 18447v2 Announce Type: replace Abstract: Existing approaches to reward inference typically assume that humans provide demonstrations according to specific behavior models.
The paper introduces a novel framework that combines vision‑language model (VLM) generated preferences with the Plackett‑Luce (PL) ranking model for reward learning in reinforcement learning. Unlike traditional pairwise Bradley‑Terry approaches, the PL formulation allows listwise rankings of multiple candidates, enabling the use of different ranking sizes (K = 3, 4, 5). Experiments on Meta‑World manipulation tasks show that PL‑based reward models train robotic policies as effectively as, or better than, pairwise, K‑wise, and RL‑VLM‑F baselines, achieving up to an 86% mean final success rate and matching the Oracle baseline on the Drawer Open task.
arXiv:2606. 24515v1 Announce Type: new Abstract: Computer-Use Agents (CUAs) execute high-level user goals by perceiving and acting directly within graphical user interfaces.