# RLMM-Flow: A Flow-based Mobile Manipulation Framework with Latent-Space Reinforcement Learning

arXiv: https://arxiv.org/abs/2607.26460 · HTML: https://mlsanity.com/paper/2607.26460/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 21 of its 30 references that are in the map.
- First posted on arXiv 29 Jul 2026.

## Placed in

- 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, 21 of 21)

- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning](https://mlsanity.com/paper/1910.00177/) (2019)
- [Point Transformer V2: Grouped Vector Attention and Partition-based Pooling](https://mlsanity.com/paper/2210.05666/) (2022)
- [IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies](https://mlsanity.com/paper/2304.10573/) (2023)
- [Integrated Task and Motion Planning](https://mlsanity.com/paper/2010.01083/) (2020)
- [3D Diffuser Actor: Policy Diffusion with 3D Scene Representations](https://mlsanity.com/paper/2402.10885/) (2024)
- [Diffusion-based Generation, Optimization, and Planning in 3D Scenes](https://mlsanity.com/paper/2301.06015/) (2023)
- [Steering Your Diffusion Policy with Latent Space Reinforcement Learning](https://mlsanity.com/paper/2506.15799/) (2025)
- [Learning a Diffusion Model Policy from Rewards via Q-Score Matching](https://mlsanity.com/paper/2312.11752/) (2023)
- [SE(3)-DiffusionFields: Learning smooth cost functions for joint grasp and motion optimization through diffusion](https://mlsanity.com/paper/2209.03855/) (2022)
- [PhyScene: Physically Interactable 3D Scene Synthesis for Embodied AI](https://mlsanity.com/paper/2404.09465/) (2024)
- [Policy Representation via Diffusion Probability Model for Reinforcement Learning](https://mlsanity.com/paper/2305.13122/) (2023)
- [FlowPolicy: Enabling Fast and Robust 3D Flow-based Policy via Consistency Flow Matching for Robot Manipulation](https://mlsanity.com/paper/2412.04987/) (2024)
- [Learning Robotic Manipulation Policies from Point Clouds with Conditional Flow Matching](https://mlsanity.com/paper/2409.07343/) (2024)
- [Motion Planning around Obstacles with Convex Optimization](https://mlsanity.com/paper/2205.04422/) (2022)
- [M2Diffuser: Diffusion-based Trajectory Optimization for Mobile Manipulation in 3D Scenes](https://mlsanity.com/paper/2410.11402/) (2024)
- [DNAct: Diffusion Guided Multi-Task 3D Policy Learning](https://mlsanity.com/paper/2403.04115/) (2024)
- [AC-DiT: Adaptive Coordination Diffusion Transformer for Mobile Manipulation](https://mlsanity.com/paper/2507.01961/) (2025)
- [Reinforcement Learning for Flow-Matching Policies](https://mlsanity.com/paper/2507.15073/) (2025)
- [M3Bench: Benchmarking Whole-body Motion Generation for Mobile Manipulation in 3D Scenes](https://mlsanity.com/paper/2410.06678/) (2024)
- [Lagrangian Perturbation Diffusion Steering: Latent Reinforcement Learning for Generative Policies](https://mlsanity.com/paper/2606.01151/) (2026)

## Abstract (arXiv)

Mobile manipulation requires generating whole-body action chunks that jointly satisfy goal reaching, collision avoidance, base kinematic constraints, manipulator joint limits, and trajectory smoothness. Flow-based generative policies provide an efficient paradigm for learning multimodal and temporally consistent motion priors from expert demonstrations, but imitation-only training cannot improve policy quality beyond the demonstration distribution. We propose RLMM-Flow, a flow-based mobile manipulation framework that combines expert flow-policy pretraining with latent-space reinforcement learning post-training. The framework first learns a flow policy that captures a multimodal whole-body motion prior from expert demonstrations. The pretrained flow policy is then frozen, while a latent steering network steers its initial noise toward higher-value action chunks. To stabilize high-dimensional latent optimization, we warm up an action-space critic before jointly training the latent critic and latent actor, and introduce coarse-to-fine latent steering that progressively expands control from a horizon-shared latent representation to a full-dimensional residual representation. Experiments on mobile manipulation motion-planning benchmarks show that RLMM-Flow substantially improves task success, collision avoidance, and trajectory quality over imitation-only flow policies and existing reinforcement learning post-training baselines, while preserving fast flow-based inference.