Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control

2026-03-18

AI Summary

This work addresses limitations in existing diffusion- or flow-matching-based imitation learning approaches for robotics, which rely on fixed inference timesteps and lack mechanisms for out-of-distribution (OOD) state detection. To overcome these issues, the authors propose Generative Control as Optimization (GeCO), reframing action generation as an iterative optimization process. GeCO learns a time-unconditional static velocity field over action sequence space via flow matching, rendering expert behaviors as stable attractors. This formulation enables adaptive allocation of computational resources according to task complexity and provides a training-free OOD detection signal through the norm of the velocity field. Integrated as a plug-and-play module into pi0-family vision-language-action (VLA) models, GeCO significantly improves task success rates, inference efficiency, and deployment safety on standard simulation benchmarks.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Search and Optimization: Learning to Search
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Diffusion models and flow matching have become a cornerstone of robotic imitation learning, yet they suffer from a structural inefficiency where inference is often bound to a fixed integration schedule that is agnostic to state complexity. This paradigm forces the policy to expend the same computational budget on trivial motions as it does on complex tasks. We introduce Generative Control as Optimization (GeCO), a time-unconditional framework that transforms action synthesis from trajectory integration into iterative optimization. GeCO learns a stationary velocity field in the action-sequence space where expert behaviors form stable attractors. Consequently, test-time inference becomes an adaptive process that allocates computation based on convergence--exiting early for simple states while refining longer for difficult ones. Furthermore, this stationary geometry yields an intrinsic, training-free safety signal, as the field norm at the optimized action serves as a robust out-of-distribution (OOD) detector, remaining low for in-distribution states while significantly increasing for anomalies. We validate GeCO on standard simulation benchmarks and demonstrate seamless scaling to pi0-series Vision-Language-Action (VLA) models. As a plug-and-play replacement for standard flow-matching heads, GeCO improves success rates and efficiency with an optimization-native mechanism for safe deployment. Video and code can be found at https://hrh6666.github.io/GeCO/

Problem

• robotic imitation learning
• flow matching
• computational efficiency
• out-of-distribution detection
• adaptive inference

Innovation

• Generative Control as Optimization
• time-unconditional flow matching
• adaptive inference
• stationary velocity field
• out-of-distribution detection