QGF (Q-Guided Flow) is a test-time reinforcement learning method that utilizes critic gradients to guide pre-trained flow-based policies toward higher-value actions, enhancing stability and scalability in continuous control.</p>\n","updatedAt":"2026-06-10T02:07:52.852Z","author":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","fullname":"taesiri","name":"taesiri","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":314,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9198866486549377},"editors":["taesiri"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.11087","authors":[{"_id":"6a28c6e0e7d78ea7587e534c","name":"Zhiyuan Zhou","hidden":false},{"_id":"6a28c6e0e7d78ea7587e534d","name":"Andy Peng","hidden":false},{"_id":"6a28c6e0e7d78ea7587e534e","name":"Charles Xu","hidden":false},{"_id":"6a28c6e0e7d78ea7587e534f","name":"Qiyang Li","hidden":false},{"_id":"6a28c6e0e7d78ea7587e5350","name":"Tobias Springenberg","hidden":false},{"_id":"6a28c6e0e7d78ea7587e5351","name":"Kevin Frans","hidden":false},{"_id":"6a28c6e0e7d78ea7587e5352","name":"Sergey Levine","hidden":false}],"publishedAt":"2026-06-09T00:00:00.000Z","submittedOnDailyAt":"2026-06-10T00:00:00.000Z","title":"Test-Time Gradient Guidance of Flow Policies in Reinforcement Learning","submittedOnDailyBy":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user","name":"taesiri"},"summary":"Expressive continuous control policies, such as diffusion and flow models, form the backbone of recent advances in scaling imitation learning for simulated and real robot control. While they are known to scale stably in the supervised imitation learning setting, incorporating them into reinforcement learning (RL) pipelines for policy improvement has proven more difficult. It often requires specialized training objectives or backpropagating through denoising processes, which cause well-known issues with stability and affect scalability. In this paper we study the question of whether simple policy improvement schemes at test time alone, leaving stable supervised policy training intact, can be a competitive alternative which sidesteps these issues. To this end, we propose QGF (Q-Guided Flow), an RL algorithm that performs policy optimization entirely at test time. QGF works by pre-training both a reference flow policy (via a standard behavioral cloning objective) and a value function critic and, at test time, using the value gradient to guide the reference policy to generate higher-value actions without any additional policy learning. Empirically, QGF outperforms prior test-time RL methods on single-task and goal-conditioned offline RL benchmarks with high-dimensional action spaces, and is competitive with state-of-the-art training-time algorithms while being much cheaper to run. Moreover, it exhibits favorable scaling with model size by avoiding the instability of actor-critic training, offering a practical and effective alternative RL algorithm with expressive policies.","upvotes":1,"discussionId":"6a28c6e1e7d78ea7587e5353","projectPage":"https://q-guided-flow.github.io/","githubRepo":"https://github.com/zhouzypaul/qgf","githubRepoAddedBy":"user","ai_summary":"QGF is an RL algorithm that improves policies at test time by using a value gradient to guide a pre-trained flow policy, avoiding training-time instability while maintaining competitive performance.","ai_keywords":["diffusion models","flow models","imitation learning","reinforcement learning","policy improvement","behavioral cloning","value function critic","test-time RL","Q-Guided Flow","action generation","policy optimization"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":17},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.11087.md"}">
Test-Time Gradient Guidance of Flow Policies in Reinforcement Learning
Abstract
QGF is an RL algorithm that improves policies at test time by using a value gradient to guide a pre-trained flow policy, avoiding training-time instability while maintaining competitive performance.
Expressive continuous control policies, such as diffusion and flow models, form the backbone of recent advances in scaling imitation learning for simulated and real robot control. While they are known to scale stably in the supervised imitation learning setting, incorporating them into reinforcement learning (RL) pipelines for policy improvement has proven more difficult. It often requires specialized training objectives or backpropagating through denoising processes, which cause well-known issues with stability and affect scalability. In this paper we study the question of whether simple policy improvement schemes at test time alone, leaving stable supervised policy training intact, can be a competitive alternative which sidesteps these issues. To this end, we propose QGF (Q-Guided Flow), an RL algorithm that performs policy optimization entirely at test time. QGF works by pre-training both a reference flow policy (via a standard behavioral cloning objective) and a value function critic and, at test time, using the value gradient to guide the reference policy to generate higher-value actions without any additional policy learning. Empirically, QGF outperforms prior test-time RL methods on single-task and goal-conditioned offline RL benchmarks with high-dimensional action spaces, and is competitive with state-of-the-art training-time algorithms while being much cheaper to run. Moreover, it exhibits favorable scaling with model size by avoiding the instability of actor-critic training, offering a practical and effective alternative RL algorithm with expressive policies.
Community
QGF (Q-Guided Flow) is a test-time reinforcement learning method that utilizes critic gradients to guide pre-trained flow-based policies toward higher-value actions, enhancing stability and scalability in continuous control.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2606.11087 in a model README.md to link it from this page.
Cite arxiv.org/abs/2606.11087 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.11087 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.