From Foundation to Application: Improving VLA Models in Practice
Summary
This paper presents LingBot-VLA 2.0, which enhances VLA foundation models for robotics by improving generalization across tasks and embodiments, expanding action space to whole-body degrees of freedom, and incorporating predictive dynamics modeling for better temporal reasoning.
View Cached Full Text
Cached at: 07/09/26, 07:34 AM
Paper page - From Foundation to Application: Improving VLA Models in Practice
Source: https://huggingface.co/papers/2607.06403 Published on Jul 7
·
Submitted byhttps://huggingface.co/Weiww99
Wei Wuon Jul 8
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
LingBot-VLA 2.0 enhances generalization across tasks and embodiments through expanded data preprocessing and training on diverse robot configurations, extends action space to include whole-body degrees of freedom for complex manipulation tasks, and incorporates predictive dynamics modeling using video representation and depth estimation for improved temporal reasoning.
Despite recent progress ofVLA foundation models, the disparity between laboratory conditions and real-world applications continues to impede their practical implementation. To bridge this gap, we present LingBot-VLA 2.0, which advances LingBot-VLA through improvements in three functional domains. (1) Generalization across tasks and embodiments. Compared to the previous version, we revamp thedata processing pipelineand curate around 60,000 hours of data for pretraining, including 50,000 hours ofrobot trajectoriesspanning 20 robot configurations and 10,000 hours ofegocentric human videos. (2) Expandedaction spacein addition to dual-arm hardware platforms. In particular, our system accommodatesdegrees of freedomfor the heads, waists, mobile bases, and dexterous hands, thereby empowering the robots to tackle more complex tasks in practical scenarios. (3)Predictive dynamics modelingfor improvedtemporal reasoning. Specifically, we formulate future prediction as a proxy task, facilitated by avideo representation modelfor semantic priors and adepth estimation modelfor geometric cues. Evaluations on theGM-100 benchmark, conducted in a generalist setting, validate the beneficial impact of these proposed modifications. Furthermore, benefiting from the expanded pretraining data that covers whole-bodydegrees of freedom, LingBot-VLA-2.0 demonstrates strongcross-embodiment long-horizon mobile manipulationcapability across the two robotic platforms.
View arXiv pageView PDFProject pageGitHub306Add to collection
Get this paper in your agent:
hf papers read 2607\.06403
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2607.06403 in a model README.md to link it from this page.
Datasets citing this paper1
#### cy0307/awesome-egocentric-atlas Viewer• Updated30 minutes ago • 816 • 1.21k • 5
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2607.06403 in a Space README.md to link it from this page.
Collections including this paper1
Similar Articles
@robbyant_brain: LingBot-VLA 2.0 is now open-source — our next-gen embodied foundation model. 60,000 hours of high-quality pretraining d…
LingBot-VLA 2.0, an open-source embodied foundation model, has been released with 60,000 hours of pretraining data supporting 20 robot configurations across 17 brands, capable of sub-130ms inference on RTX 4090.
InternVLA-A1.5: Unifying Understanding, Latent Foresight, and Action for Compositional Generalization
InternVLA-A1.5 integrates pretrained vision-language models with future prediction in latent space to enable efficient robot manipulation with compositional generalization and long-horizon execution, achieving state-of-the-art results on simulation benchmarks.
LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories
LabVLA is a vision-language-action model for scientific laboratory automation, trained with a two-stage approach combining action token pretraining and flow matching. It achieves state-of-the-art success rates on the LabUtopia benchmark by leveraging simulated data to bridge the gap between household demonstrations and lab-specific tasks.
AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding
AffordanceVLA introduces a unified framework using structured affordance forecasting as an intermediate representation to improve perception-action mapping in robotic manipulation, leveraging vision-language models and a Mixture-of-Transformer architecture.
IntentVLA: Short-Horizon Intent Modeling for Aliased Robot Manipulation
IntentVLA is a history-conditioned visual-language-action framework that improves robot imitation learning stability by encoding short-horizon intents from visual observations, addressing challenges from partial observability and ambiguous observations. It also introduces AliasBench, an ambiguity-aware benchmark for evaluating such methods.