AI & ML
RAVEL: Asynchronous Rolling Inference for Flow-Based Vision-Language-Action Models
arXiv cs.RO0 views
arXiv:2609.34170v2 Announce Type: replace Abstract: Flow-based vision-language-action (VLA) models are highly effective for generalist robot manipulation, yet their reliance on computationally expensive VLM encoding and multi-step iterative action generation imposes a significant latency bottleneck. The resulting inference latency makes it difficult for robots to respond quickly, especially in dynamic environments. We address this limitation with RAVEL (Rolling Asynchronous VLA Enabling Low-Latency Control), an asynchronous inference framework that addresses the computational bottlenecks of both the VLM backbone and the action expert. To reduce the delay from multi-step action denoising, RAVEL allows near-term actions to be executed after a single additional denoising step by carrying partially denoised future actions forward in a rolling buffer. To avoid blocking on slow VLM encoding, RAVEL decouples VLM encoding from rolling action generation, allowing the action expert to operate continuously using the latest available VLM context, while a lightweight Fast Observation Pathway (FOP) directly conditions the action expert on current observations. Across simulated and real-world manipulation tasks, RAVEL consistently achieves substantially lower response latency while maintaining the task capability of the underlying VLA, enabling high-frequency and responsive closed-loop control.
Source:https://arxiv.org/abs/2609.34170PreviousGLoTouch: Global-to-Local Haptic Perception Using a Parallel Gripper for Object Search, Recognition, and Grasping Without External VisionNextVideo2SwimFish: An Automated Pipeline for Reconstructing Controllable Fish Models and Biological Locomotion from Real Fish Videos
Related posts
Mixing Regularization for Average-Reward $Q$-Learning and $Q$-Planning
0arXiv stat.ML
Pragmatic DML with AI-Learned Representations
0arXiv stat.ML
Arbitrary-Accuracy Neural Approximation with Optimal Neuron Count and Optimal Bit Complexity
0arXiv stat.ML
Prior-Amortized In-Context Bayesian Inference for Generalized Linear Mixed-Effects Models
0arXiv stat.ML
Agent Meltdowns: The Road to Hell Is Paved with Helpful Agents
0arXiv cs.CR
Composable Driving Style for End-to-End Autonomous Driving
0arXiv cs.RO