# dVLA-RL: Reinforcement Learning over Denoising Trajectories for Discrete Diffusion Vision-Language-Action Models

arXiv: https://arxiv.org/abs/2606.23623 · HTML: https://mlsanity.com/paper/2606.23623/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Efficient VLA inference and adaptation, one of 13 directions in Vision-language-action models.
- Builds on 30 of its 37 references that are in the map.
- First posted on arXiv 22 Jun 2026.

## Placed in

- Vision-language-action models / [Efficient VLA inference and adaptation](https://mlsanity.com/vla/efficient-vla-inference-and-adaptation/)
- Diffusion models / [Diffusion and flow policies for robotics](https://mlsanity.com/diffusion/diffusion-and-flow-policies-for-robotics/)
- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 30 of 30)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Structured Denoising Diffusion Models in Discrete State-Spaces](https://mlsanity.com/paper/2107.03006/) (2021)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [Large Language Diffusion Models](https://mlsanity.com/paper/2502.09992/) (2025)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [Simple and Effective Masked Diffusion Language Models](https://mlsanity.com/paper/2406.07524/) (2024)
- [Flow-GRPO: Training Flow Matching Models via Online RL](https://mlsanity.com/paper/2505.05470/) (2025)
- [Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions](https://mlsanity.com/paper/2102.05379/) (2021)
- [Dream 7B: Diffusion Large Language Models](https://mlsanity.com/paper/2508.15487/) (2025)
- [UniVLA: Learning to Act Anywhere with Task-centric Latent Actions](https://mlsanity.com/paper/2505.06111/) (2025)
- [Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning](https://mlsanity.com/paper/2601.16163/) (2026)
- [Motus: A Unified Latent Action World Model](https://mlsanity.com/paper/2512.13030/) (2025)
- [MMaDA: Multimodal Large Diffusion Language Models](https://mlsanity.com/paper/2505.15809/) (2025)
- [DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation](https://mlsanity.com/paper/2506.20639/) (2025)
- [d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learning](https://mlsanity.com/paper/2504.12216/) (2025)
- [Flow Q-Learning](https://mlsanity.com/paper/2502.02538/) (2025)
- [VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning](https://mlsanity.com/paper/2505.18719/) (2025)
- [SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning](https://mlsanity.com/paper/2509.09674/) (2025)
- [LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning](https://mlsanity.com/paper/2505.16933/) (2025)
- [Interactive Post-Training for Vision-Language-Action Models](https://mlsanity.com/paper/2505.17016/) (2025)
- [Discrete Diffusion VLA: Bringing Discrete Diffusion to Action Decoding in Vision-Language-Action Policies](https://mlsanity.com/paper/2508.20072/) (2025)
- [Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models](https://mlsanity.com/paper/2509.06949/) (2025)
- [TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimization](https://mlsanity.com/paper/2506.08440/) (2025)
- [RL Token: Bootstrapping Online RL with Vision-Language-Action Models](https://mlsanity.com/paper/2604.23073/) (2026)
- [Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Process](https://mlsanity.com/paper/2511.01718/) (2025)
- [dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought](https://mlsanity.com/paper/2509.25681/) (2025)
- [MM-ACT: Learn from Multimodal Parallel Generation to Act](https://mlsanity.com/paper/2512.00975/) (2025)
- [MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation](https://mlsanity.com/paper/2603.25406/) (2026)

## Built on by (most cited first, 1 of 1)

- [Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model](https://mlsanity.com/paper/2609.13053/) (2026)

## Abstract (arXiv)

Vision-Language-Action (VLA) models have established a powerful paradigm for generalist robotic manipulation by grounding control into the semantic reasoning of VLMs. Prevailing architectures typically model actions continuously via diffusion or flow processes, or discretely through either autoregressive generation or parallel decoding. Recently, Discrete Diffusion VLAs (dVLAs) have emerged as a distinct alternative, unifying vision, language, and action into a single discrete token space via masked generative modeling. While combining iterative refinement with unified representations, its training has thus far been restricted to Supervised Fine-Tuning (SFT), leaving the potential of Reinforcement Learning (RL) for further policy refinement largely unexplored. A fundamental challenge in RL for dVLAs is that the marginal probability of the final action generated by dVLAs remains intractable. To solve this problem, we propose dVLA-RL, shifting the learning objective from the marginal action probability to the joint probability of the sampled generation path. Specifically, by modeling the denoising process as a Markov Decision Process (MDP), we mathematically formulate this path probability as a product of step-wise transitions. This trajectory-level objective provides a unified formulation that natively accommodates variable denoising steps. Leveraging this intrinsic fexibility, we introduce a unified step scheduling approach for complex multi-task learning, tailoring denoising steps to specific task complexities to maximize both success rates and computational effciency. Extensive evaluations demonstrate that our approach achieves a success rate of 99.7% on LIBERO. Furthermore, it establishes strong VLA-based results on RoboTwin 2.0 by delivering a 30.6% improvement over the SFT baseline, remaining competitive with strong World-Action Model baselines.