Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization

• URL: http://arxiv.org/abs/2602.09722v1
• Date: Tue, 10 Feb 2026 12:25:43 GMT
• Title: Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization
• Abstract summary: Vision-Language-Action (VLA) models show strong promise for generalist robot control.<n>It remains unclear whether -- and under what conditions -- the standard "scale data" recipe translates to robotics.<n>We present a systematic, controlled study of VLA scaling that revisits core training choices for pretraining across diverse robots.
• Score: 65.37179698521766
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While Vision-Language-Action (VLA) models show strong promise for generalist robot control, it remains unclear whether -- and under what conditions -- the standard "scale data" recipe translates to robotics, where training data is inherently heterogeneous across embodiments, sensors, and action spaces. We present a systematic, controlled study of VLA scaling that revisits core training choices for pretraining across diverse robots. Using a representative VLA framework that combines a vision-language backbone with flow-matching, we ablate key design decisions under matched conditions and evaluate in extensive simulation and real-robot experiments. To improve the reliability of real-world results, we introduce a Grouped Blind Ensemble protocol that blinds operators to model identity and separates policy execution from outcome judgment, reducing experimenter bias. Our analysis targets three dimensions of VLA scaling. (1) Physical alignment: we show that a unified end-effector (EEF)-relative action representation is critical for robust cross-embodiment transfer. (2) Embodiment mixture: we find that naively pooling heterogeneous robot datasets often induces negative transfer rather than gains, underscoring the fragility of indiscriminate data scaling. (3) Training regularization: we observe that intuitive strategies, such as sensory dropout and multi-stage fine-tuning, do not consistently improve performance at scale. Together, this study challenge some common assumptions about embodied scaling and provide practical guidance for training large-scale VLA policies from diverse robotic data. Project website: https://research.beingbeyond.com/rethink_vla

Related papers

• How to Better Train VLAs: Lessons Learned From the REAL-I Challenge at ICRA 2026 [19.591059484557256]
  The first Real-world Embodied AI Learning (REAL-I) Challenge at ICRA 2026 examined this question through simulation, real-robot evaluation, and an on-site final on a shared dual-arm humanoid platform.<n>We describe the challenge tasks, data and deployment interfaces, and competition results, then compare the approaches contributed by NUS-CLEAR, RCL-Lab, and Deeptouch.ai.<n>Their systems combined pretrained vision-language-action models and task-specific imitation policies with different strategies for data curation, staged adaptation, checkpoint selection, and action-space design.
  arXiv  Detail & Related papers  (2026-09-12T03:12:46Z)
• VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment [17.204682551730205]
  We present VLAFlow, a unified flow-matching framework for comparison of VLA training objectives.<n>We evaluate four paradigms under the same pi0-style architecture, shared VLM backbone, action expert, and 14-dimensional action space.
  arXiv  Detail & Related papers  (2026-07-02T01:38:16Z)
• Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data [93.7685703383343]
  Vision-language models (VLMs) are powerful general-purpose reasoners, yet converting them into robot control policies is surprisingly difficult.<n>We argue that this gap can be bridged gradually with the right intermediate data.<n>We show that mixing task-relevant out-of-distribution ETC data with a small amount of action data enables the model to generalize to novel visual-language conditions.
  arXiv  Detail & Related papers  (2026-06-07T08:57:51Z)
• VISTA: Vision-Grounded and Physics-Validated Adaptation of UMI data for VLA Training [52.05483137072975]
  Universal Manipulation Interface (UMI) enables scalable real-world robot data collection without hardware-specific teleoperation.<n>We present VISTA, a framework that bridges this dual gap through three synergistic components.<n>We release the physical-validation pipeline, UMI-VQA, validated trajectory data, and the pre-trained model for the community.
  arXiv  Detail & Related papers  (2026-06-03T10:38:45Z)
• Joint-Aligned Latent Action: Towards Scalable VLA Pretraining in the Wild [37.43820830488286]
  We present JALA, a pretraining framework that learns Jointly-Aligned Latent Actions.<n>We scale this approach with UniHand-Mix, a 7.5M video corpus (>2,000 hours) blending laboratory and in-the-wild footage.<n>Experiments demonstrate that JALA generates more realistic hand motions in both controlled and unconstrained scenarios.
  arXiv  Detail & Related papers  (2026-02-25T09:46:42Z)
• Self-Correcting VLA: Online Action Refinement via Sparse World Imagination [55.982504915794514]
  We propose Self-Correcting VLA (SC-VLA), which achieve self-improvement by intrinsically guiding action refinement through sparse imagination.<n>SC-VLA achieve state-of-the-art performance, yielding the highest task throughput with 16% fewer steps and a 9% higher success rate than the best-performing baselines.
  arXiv  Detail & Related papers  (2026-02-25T06:58:06Z)
• ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning [31.000965640377128]
  ABot-M0 is a framework that builds a systematic data curation pipeline.<n>It enables end-to-end transformation of heterogeneous raw data into unified, efficient representations.<n>ABot-M0 supports modular perception via a dual-stream mechanism.
  arXiv  Detail & Related papers  (2026-02-11T16:47:01Z)
• LIBERO-X: Robustness Litmus for Vision-Language-Action Models [32.29541801424534]
  This work systematically rethinks VLA benchmarking from both evaluation and data perspectives.<n>We introduce LIBERO-X, a benchmark featuring a hierarchical evaluation protocol with progressive difficulty levels targeting three core capabilities.<n> Experiments with representative VLA models reveal significant performance drops under cumulative perturbations.
  arXiv  Detail & Related papers  (2026-02-06T09:59:12Z)
• Reshaping Action Error Distributions for Reliable Vision-Language-Action Models [69.38615670891038]
  In robotic manipulation, vision-language-action (VLA) models have emerged as a promising paradigm for learning generalizable and scalable robot policies.<n>We focus on continuous-action VLA models and move beyond conventional MSE-based regression by reshaping action error distributions during training.<n>We evaluate our approaches across standard, few-shot, and noisy settings on multiple representative VLA architectures.
  arXiv  Detail & Related papers  (2026-02-04T05:37:09Z)
• Steering Vision-Language-Action Models as Anti-Exploration: A Test-Time Scaling Approach [78.4812458793128]
  We propose textbfTACO, a test-time-scaling framework that applies a lightweight pseudo-count estimator as a high-fidelity verifier of action chunks.<n>Our method resembles the classical anti-exploration principle in offline reinforcement learning (RL), and being gradient-free, it incurs significant computational benefits.
  arXiv  Detail & Related papers  (2025-12-02T14:42:54Z)
• Align-Then-stEer: Adapting the Vision-Language Action Models through Unified Latent Guidance [63.33213516925946]
  We introduce textbfAlign-Then-stEer (textttATE), a novel, data-efficient, and plug-and-play adaptation framework.<n>Our work presents a general and lightweight solution that greatly enhances the practicality of deploying VLA models to new robotic platforms and tasks.
  arXiv  Detail & Related papers  (2025-09-02T07:51:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.