Q-VGM: Q-Guided Value-Gradient Matching for Flow-Matching VLA Policies

• URL: http://arxiv.org/abs/2606.08015v1
• Date: Sat, 06 Jun 2026 07:10:25 GMT
• Title: Q-VGM: Q-Guided Value-Gradient Matching for Flow-Matching VLA Policies
• Abstract summary: We propose Q-Guided Value-Gradient Matching (Q-VGM) as an off-policy reinforcement learning (RL) method.<n>Q-VGM sidesteps issues by leveraging VGG-Flow, a value-gradient view of flow alignment in generative modeling.<n>On LIBERO, Q-VGM raises the average success rate from 75.0% to 92.5%; on RoboTwin 2.0, from 76.4% to 87.2%; and on two real-robot tabletop tasks, from 40.0% to 67.5%.
• Score: 14.519898493996891
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose Q-Guided Value-Gradient Matching (Q-VGM), an off-policy reinforcement learning (RL) method that tackles a long-standing challenge in fine-tuning flow-matching vision-language-action (VLA) policies: efficiently improving an expressive flow-matching action expert with respect to a learned Q-function. Effective improvement must exploit the first-order (gradient) information of the critic, but this is difficult for flow policies, because directly back-propagating the value through their multi-step denoising process is numerically unstable at VLA scale, while the tractable action likelihoods required by policy-gradient methods are unavailable under iterative denoising. Existing value-based methods either backpropagate through the full denoising chain, use the critic only at test time without updating the policy, or distill critic-improved actions as terminal labels without supervising the velocity field. Q-VGM sidesteps these issues by leveraging VGG-Flow, a value-gradient view of flow alignment in generative modeling that transforms value gradient into a denoising-time value-gradient field rather than an unstable end-to-end objective. This requires no action likelihoods and no backpropagation through the denoising chain, and operates on a fixed replay buffer. The critic is an action-sensitive Cal-QL ensemble over compact RLT features with per-layer action injection. Q-VGM enables a practical few-shot initialization then learn-from-experience paradigm: starting from a few-shot-SFT pi0.5 VLA, the method leverages self-generated rollout data to substantially improve task performance without additional expert supervision. On LIBERO, Q-VGM raises the average success rate from 75.0% to 92.5%; on RoboTwin 2.0, from 76.4% to 87.2%; and on two real-robot tabletop tasks, from 40.0% to 67.5%, outperforming all same-backbone, same-critic baselines across all three settings.

Related papers

• VINE: Taming Generative Control Policies for Reinforcement Learning [13.904046574218567]
  Flow-matching policies have emerged as an effective policy parameterization for robot learning.<n>Prior works observed that scaling these policies with value-gradient reinforcement learning (RL) often leads to training instability.<n>We propose VINE, an RL-oriented sampling method that enables stable end-to-end value-gradient optimization for flow-matching policies.
  arXiv  Detail & Related papers  (2026-07-11T15:45:12Z)
• FORCE: Efficient VLA Reinforcement Fine-Tuning via Value-Calibrated Warm-up and Self-Distillation [40.53898394579726]
  FORCE is a 3-stage framework that stabilizes fine-tuning by tackling both issues.<n>FORCE achieves a 79% absolute improvement in success rates and outperform prior RL methods by 10%, while accelerating training by 32.5%.
  arXiv  Detail & Related papers  (2026-06-24T16:23:18Z)
• VIMPO: Value-Implicit Policy Optimization for LLMs [106.88933849641272]
  Group-relative methods such as GRPO avoid training a critic, but typically assign a trajectory-level advantage to every token.<n>Actor-critic methods provide denser learning signals, but require a learned value function with its own training instability.<n>We introduce VIMPO, a critic-free policy optimization method that derives a policy-implied value function from optimality conditions of KL-regularized reinforcement learning.
  arXiv  Detail & Related papers  (2026-06-18T09:44:12Z)
• QPILOTS: Efficient Test-Time Q-Steering for Flow Policies [20.217020870532686]
  QPILOTS is a method that leaves the original policy unmodified and steers the denoising process at inference time.<n>On a standard offline-to-online RL benchmark, QPILOTS achieves the best aggregate performance, reaching an average success rate of 90% across 50 tasks.
  arXiv  Detail & Related papers  (2026-06-11T18:22:03Z)
• Test-Time Gradient Guidance of Flow Policies in Reinforcement Learning [50.738952715864116]
  Expressive continuous control policies form the backbone of advances in scaling imitation learning for simulated and real robot control.<n>We propose QGF (Q-Guided Flow), an RL algorithm that performs policy optimization entirely at test time.<n> Empirically, QGF outperforms prior test-time RL methods on single-task and goal-conditioned offline RL benchmarks.
  arXiv  Detail & Related papers  (2026-06-09T16:45:57Z)
• Denoising Tells When to Replan: Denoising-Variance Adaptive Chunking for Flow-Based Robot Policies [18.476299874941592]
  Action chunking has become a common inference strategy for flow-based robot policies.<n>We propose DVAC, a test-time method that adaptively determines how many actions to execute from each predicted chunk.<n>Experiments on LIBERO, RoboTwin, CALVIN, and real-world manipulation show that DVAC improves task success while reducing replanning frequency.
  arXiv  Detail & Related papers  (2026-06-02T16:26:32Z)
• Q-learning with Adjoint Matching [58.78551025170267]
  We propose Q-learning with Adjoint Matching (QAM), a novel TD-based reinforcement learning (RL) algorithm.<n>QAM sidesteps two challenges by leveraging adjoint matching, a recently proposed technique in generative modeling.<n>It consistently outperforms prior approaches on hard, sparse reward tasks in both offline and offline-to-online RL.
  arXiv  Detail & Related papers  (2026-01-20T18:45:34Z)
• TTF-VLA: Temporal Token Fusion via Pixel-Attention Integration for Vision-Language-Action Models [29.878993349922368]
  Vision-Language-Action (VLA) models process visual inputs independently at each timestep, discarding valuable temporal information inherent in robotic manipulation tasks.<n>We propose Temporal Token Fusion (TTF), a training-free approach that integrates historical and current visual representations to enhance VLA inference quality.
  arXiv  Detail & Related papers  (2025-08-15T12:03:34Z)
• Relative Entropy Pathwise Policy Optimization [66.03329137921949]
  We present an on-policy algorithm that trains Q-value models purely from on-policy trajectories.<n>We show how to combine policies for exploration with constrained updates for stable training, and evaluate important architectural components that stabilize value function learning.
  arXiv  Detail & Related papers  (2025-07-15T06:24:07Z)
• A Perspective of Q-value Estimation on Offline-to-Online Reinforcement Learning [54.48409201256968]
  offline-to-online Reinforcement Learning (O2O RL) aims to improve the performance of offline pretrained policy using only a few online samples. Most O2O methods focus on the balance between RL objective and pessimism, or the utilization of offline and online samples.
  arXiv  Detail & Related papers  (2023-12-12T19:24:35Z)
• Offline Reinforcement Learning with Implicit Q-Learning [85.62618088890787]
  Current offline reinforcement learning methods need to query the value of unseen actions during training to improve the policy. We propose an offline RL method that never needs to evaluate actions outside of the dataset. This method enables the learned policy to improve substantially over the best behavior in the data through generalization.
  arXiv  Detail & Related papers  (2021-10-12T17:05:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.