Recompose and Refine Latent Reasoning Flows for Vision-Language-Action Models
On the map: Vision-language-action models → Reasoning and chain-of-thought in VLA
Abstract (arXiv)
Latent reasoning enables vision-language-action (VLA) models to transform multimodal observations into task-relevant internal states before generating continuous robot actions. While existing methods learn to generate or refine such states for each policy query, they discard successful reasoning after execution and therefore reconstruct similar computation from scratch. We present Reasoning and Flow Memory (FLOWMEM), a unified VLA model that turns successful latent computation into reusable reasoning experience. Rather than appending a fixed retrieved context, FLOWMEM dynamically retrieves and recomposes compatible latent fragments as the embodied context evolves, forming a reasoning route that follows the temporal structure and progress of successful computation. The route is then refined using current visual and proprioceptive evidence before it conditions action generation. Experiments on RoboMME and LIBERO-Plus show that FLOWMEM attains 48.0% and 77.3% success, outperforming memory-free policies by 1.7 and 4.1 percentage points, respectively. These results demonstrate the value of reusing successful latent computation for closed-loop VLA control.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Reasoning and chain-of-thought in VLA, one of 13 directions in Vision-language-action models. Builds on 9 of its 27 references that are in the map. First posted on arXiv 8 Oct 2026.
Builds on
9 of 27 references.
Closest ancestors · 6
Same direction, newest first
- 2026Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation8 Jul
- 2026WeaveLA: Event Driven Cross-Subtask Latent Memory Weaving for Repetitive Robot Manipulation16 Jun
- 2026μVLA: On Recurrent Memory for Partially Observable Manipulation in VLA Models10 Jun
- 2026MemoryVLA++: Temporal Modeling via Memory and Imagination in Vision-Language-Action Models8 Jun
- 2026ReMem-VLA: Empowering Vision-Language-Action Model with Memory via Dual-Level Recurrent Queries13 Mar
- 2024Robotic Control via Embodied Chain-of-Thought Reasoning11 Jul
Other references · 3
- 2025SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulationcited by 61
- 2026Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoningcited by 18
- 2026TFP: Temporally Conditioned Memory-Fusion Policies for Visuomotor Learningcited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Latent Reasoning VLA: Latent Thinking and Prediction for Vision-Language-Action Models
- 2026Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning
- 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning
- 2026G0.5: One Autoregressive Stream for Robot Reasoning and Action
- 2026Memory Without Curation: Learning Multi-Stream Latent Memory for Robot Policies
- 2026Being-H0.7: A Latent World-Action Model from Egocentric Videos
- 2025Learning Affordances at Inference-Time for Vision-Language-Action Models
- 2025AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention