High-Fidelity One-Step Generative Visuomotor Policy via Recursive Correction, Frequency Consistency, and Contrastive Flow Matching

2026-07-04

AI Summary

Existing single-step generative visuomotor policies often suffer from spatial bias, loss of high-frequency details, and ambiguous action modes due to oversimplified inference. To address these limitations, this work proposes a high-fidelity single-step framework that integrates Recursive Consistent Action Flow (RCAF) to correct spatial truncation errors, Dual-Timestep Frequency Consistency (DTFC) to preserve high-frequency manipulation details, and Contrastive Flow Matching (CFM) to disentangle multimodal action distributions. Requiring only a single forward pass (1 NFE), the method achieves or surpasses the control performance of multi-step baselines—such as 10-step policies—on RoboTwin, Adroit, DexArt, and real robotic platforms, significantly reducing latency while enhancing both precision and action diversity.

Abstract

Generative models such as diffusion and flow matching have advanced robotic visuomotor policies by modeling multimodal action distributions, but their multi-step sampling or ODE solving introduces inference latency. Existing one-step acceleration methods often compress the whole generation process into a single large update, leading to spatial deviation, frequency distortion, and mode averaging. This paper proposes a high-fidelity one-step generative visuomotor policy framework that addresses these issues with three complementary mechanisms. Recursive Consistent Action Flow (RCAF) uses recursive correction to compensate for spatial truncation errors and align one-step predictions with refined flow trajectories. Dual-Timestep Frequency Consistency (DTFC) preserves high-frequency manipulation details through adaptive spectral consistency across flow timesteps. Contrastive Flow Matching (CFM) separates entangled action flows with a margin-based repulsive objective, reducing ambiguous actions in multimodal manipulation. Experiments on RoboTwin, RoboTwin 2.0, Adroit, DexArt, and real-world robot platforms show that the proposed method achieves competitive or superior performance compared with strong 10-step generative policy baselines while requiring only one forward pass (1 NFE), enabling low-latency visuomotor control.

Problem

• one-step generative policy
• spatial deviation
• frequency distortion
• mode averaging
• visuomotor control

Innovation

• Recursive Correction
• Frequency Consistency
• Contrastive Flow Matching
• One-Step Generation
• Visuomotor Policy