Let It Be Simple: One-Step Action Generation for Vision-Language-Action Models
On the map: Vision-language-action models → Efficient VLA inference and adaptation
Abstract (arXiv)
Generating diverse images from sparse text is hard; generating compact actions from rich observations is easier. From the condition-target view, Vision-Language-Action (VLA) thus aligns with image-to-text, not text-to-image. We formalize this view through the irreducible velocity loss Rv(t,c) of standard flow matching and validate it with a controlled 8-mode toy experiment and image-to-text MNIST task. We then show that high-noise training boosts one-step VLA decoding on standard LIBERO, achieving 95.6% on LIBERO-Long, and remains competitive across LIBERO-Plus, LIBERO-Pro, and real-world robot tasks, while ablations that weaken the condition or expand the horizon predictably erase the one-step gain. These results suggest that whether one-step action generation works in VLA depends not on specialized training, but on the condition-target structure.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Efficient VLA inference and adaptation, one of 13 directions in Vision-language-action models. Builds on 45 of its 54 references that are in the map. First posted on arXiv 4 Jun 2026.
Builds on
45 of 54 references.
Closest ancestors · 3
Same direction, newest first
- 2026SnapFlow: One-Step Action Generation for Flow-Matching VLAs via Progressive Self-Distillation7 Apr
- 2025CEED-VLA: Consistency Vision-Language-Action Model with Early-Exit Decoding16 Jun
- 2025FAST: Efficient Action Tokenization for Vision-Language-Action Models16 Jan
Other references · 42
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2020Denoising Diffusion Probabilistic Modelscited by 20,851
- 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
- 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
- 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
- 2022Scalable Diffusion Models with Transformerscited by 5,931
- 2022Flow Matching for Generative Modelingcited by 5,426
- 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
- 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
- 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
- 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
- 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
30 more
- 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
- 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
- 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
- 2023Consistency Modelscited by 1,778
- 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
- 2024HunyuanVideo: A Systematic Framework For Large Video Generative Modelscited by 1,597
- 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
- 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
- 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
- 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,005
- 2023One-step Diffusion with Distribution Matching Distillationcited by 978
- 2023Stochastic Interpolants: A Unifying Framework for Flows and Diffusionscited by 897
- 2024SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformerscited by 707
- 2024PaliGemma: A versatile 3B VLM for transfercited by 679
- 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 633
- 2025Mean Flows for One-step Generative Modelingcited by 524
- 2025Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Modelscited by 376
- 2024One Step Diffusion via Shortcut Modelscited by 373
- 2022MAGVIT: Masked Generative Video Transformercited by 341
- 2025LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Modelscited by 278
- 2025Diffusion Transformers with Representation Autoencoderscited by 275
- 2025LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorizationcited by 166
- 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Modelcited by 140
- 2024Flow map matching with stochastic interpolants: A mathematical framework for consistency modelscited by 99
- 2026Generative Modeling via Driftingcited by 98
- 2026LangFlow: Continuous Diffusion Rivals Discrete in Language Modelingcited by 28
- 2026SimVLA: A Simple VLA Baseline for Robotic Manipulationcited by 10
- 2025Demystifying Flux Architecturecited by 9
- 2026vla-eval: A Unified Evaluation Harness for Vision-Language-Action Modelscited by 7
- 2026A Hidden Semantic Bottleneck in Conditional Embeddings of Diffusion Transformerscited by 2
Built on by
Citing papers per year: 2026: 3.
3 papers.
Similar, not cited
Close in wording, no citation either way
- 2026StairVLA: Stage-Aware Hierarchical Action Generation for Vision-Language-Action Models
- 2026CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies
- 2026π0-EqM: Equilibrium Matching for Closed-Loop Vision-Language-Action Control
- 2026Mean-Flow based One-Step Vision-Language-Action
- 2025Discrete Diffusion VLA: Bringing Discrete Diffusion to Action Decoding in Vision-Language-Action Policies
- 2026AdaVLA: Adaptive Step Flow Matching for Training-free Acceleration of Vision-Language-Action Models
- 2026A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model
- 2026IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies