cs.RONov 15, 2026

NF-VLA: A Normalizing-Flow Baseline for Vision-Language-Action Learning

Authors: Ruichen Deng, Qinyu Zhao, Yong-Hyun Park, Mutian Tong, Liang Zheng, Jiatao Gu

Organizations: Department of Computer and Information Science University of Pennsylvania · School of Computing Australian National University

Abstract

Autoregressive normalizing flows (NFs) provide a tractable density for learning and sampling continuous robot actions. We show that a single maximum-likelihood objective can train vision--language--action (VLA) policies competitive with leading methods on manipulation benchmarks. We present NF-VLA, which combines conditional autoregressive normalizing flows with score-based action refinement. Its deep--shallow architecture concentrates visual, language, and proprioceptive prefix conditioning in a high-capacity deep block, while unconditioned shallow blocks provide a shared invertible action transformation. The vision--language backbone and flow are jointly trained by exact maximum likelihood on Gaussian-smoothed action chunks. The same learned density supplies a single score-based correction at inference, refining generated actions to account for training-time smoothing without a separately trained denoiser. Trained on LIBERO from a pretrained vision--language backbone without robot pretraining, NF-VLA achieves 98.35% success across the four standard suites and 72.6% on LIBERO-Plus using the same checkpoint without further fine-tuning. Ablations show that deep-only prefix conditioning and score-based refinement both improve closed-loop success. Together, these results establish carefully structured autoregressive flows as a competitive approach to VLA control, with one learned density supporting policy training, action generation, and refinement.

Figures & tables

Explore similar work

CardsList
  1. FAVR: Flow-Aware Visual Residual Alignment for Vision-Language-Action Policies

    Nov 16, 2026Qixuan Wu, Xiaole Yang, Yuguang Yang +5Test-Time Adaptation of VLAsTemporal Modeling for VLAs

  2. CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies

    Apr 27, 2026Fan Du, Feng Yan, Jianxiong Wu +8Robotic ControlEfficient VLA Inference

  3. IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies

    Sep 10, 2026Kian Hosseinkhani, Qinhe Peng, George Shramko +6Efficient VLA InferenceVision-Language-Action Models