Openvla
PulseAugur coverage of Openvla — every cluster mentioning Openvla across labs, papers, and developer communities, ranked by signal.
5 day(s) with sentiment data
-
New DRAP method enhances continual learning in physical AI
Researchers have introduced DRAP (Dynamics-Risk Adaptive Plasticity), a novel method designed to improve continual learning in physical AI systems. This approach addresses the challenge of policy errors accumulating ove…
-
逐际动力发布FluxVLA Engine，统一具身智能研发流程
逐际动力 (Interlink Dynamics) has released the FluxVLA Engine, an open-source engineering framework designed to streamline the development of embodied AI. This engine aims to unify data, training, evaluation, and deployment…
-
New method predicts robot task failure using VLA model activations
Researchers have developed a method to predict task failure in Vision-Language-Action (VLA) models used for robotics. By analyzing sparse autoencoder (SAE) representations of VLA activations, they can identify patterns …
-
New method attributes VLA model failures without costly rollouts
Researchers have developed a new method for failure attribution in vision-language-action (VLA) models, which are sensitive to visual corruptions. This approach estimates the benefit of removing specific corruptions wit…
-
New formulation unifies embodied foundation models for robot learning
Researchers have proposed a unified, architecture-agnostic probabilistic formulation for embodied foundation models, which are used in robot learning and control. This new formulation leverages stochastic kernels to cha…
-
New research advances vision-language-action models for robotics and reasoning · 9 sources tracked
Multiple research papers explore advancements in vision-language-action (VLA) models, focusing on improving their robustness, interpretability, and efficiency. One paper introduces JADE, a joint denoising approach that …
-
New framework WareFly-VLA enables language-guided UAV control in warehouses
Researchers have introduced WareFly-VLA, a new framework and dataset designed for vision-language-action (VLA) models to control unmanned aerial vehicles (UAVs) in smart warehouses. The framework includes a dataset of h…
-
Apple's REFACTOR-VLA learns reusable skills for robots
Apple's REFACTOR-VLA system addresses limitations in current vision-language-action (VLA) models by learning reusable skills through a wake/sleep architecture. Unlike monolithic models that output raw commands, REFACTOR…
-
SpecVLA framework enhances VLA model efficiency for embodied AI
Researchers have developed SpecVLA, a novel framework for co-designing algorithms and hardware architectures to improve the efficiency of Vision-Language-Action (VLA) models in embodied AI. This approach leverages the o…
-
New VLAGuard framework enhances robot defense against physical attention hijacking
Researchers have developed VLAGuard, a framework designed to protect Vision-Language-Action (VLA) robots operating as mobile edge nodes in wireless sensor networks from physical adversarial attacks. The framework includ…
-
Google scientists use video world models to train robots cheaply
Researchers from Google Labs and NYU have developed a novel approach to train robots by using video generation models as a substitute for real-world interaction. This method, dubbed "World Gym," allows robots to undergo…
-
AI Agents: Defining the Future of Robotics with Vision-Language-Action Models
An AI agent is defined by its ability to perceive, decide, act, and pursue a goal, distinguishing it from a simple chatbot through the use of tools and a feedback loop. This framework is crucial for developing robots ca…
-
DaXiao Robotics releases unified embodied AI model ACE-Brain-0.5
DaXiao Robotics has released ACE-Brain-0.5, a new unified embodied foundation model designed for Physical Agentic AI. This model integrates spatial perception, decision planning, embodied interaction, and self-evaluatio…
-
AI model activations signal near-term failure under visual shifts
Researchers have developed a method to detect potential failures in Vision Language Action (VLA) models like OpenVLA by analyzing their internal activations. During controlled experiments involving visual distribution s…
-
Zhipingfang secures $500M at $2B valuation, leading China's embodied AI race · 1 source tracked
Chinese embodied intelligence startup Zhipingfang has secured nearly 500 million USD in new funding, valuing the company at over 200 billion RMB. This funding round, which saw participation from national investment fund…
-
New research enhances VLA models for robotics and visual reasoning
Recent research explores enhancing Vision-Language-Action (VLA) models for robotic manipulation and general visual reasoning. Studies investigate grounding sim-to-real generalization through domain randomization and pho…
-
New VLA Models Refine Action Plans in Latent Space for Robotics
Researchers have developed new frameworks for Vision-Language-Action (VLA) models to improve robotic manipulation tasks. One approach, PearlVLA, refines action plans within the latent space of a vision-language model to…
-
New frameworks enhance AI embodied manipulation with reasoning and physics grounding · 4 sources tracked
Researchers have developed Guava, a framework designed to enhance embodied manipulation capabilities in AI agents by integrating high-level reasoning with external modules for perception, planning, and control. This har…
-
New framework detects robot execution failures using trajectory data
Researchers have developed a new framework called Hide-and-Seek to improve the reliability of robots using Vision-Language-Action (VLA) models. This method detects execution failures by identifying specific actions that…
-
New VLA-Hijack attack exploits visual self-localization in AI models
Researchers have developed VLA-Hijack, a novel adversarial framework designed to exploit vulnerabilities in Vision-Language-Action (VLA) models. This method targets the models' reliance on visual self-localization of ro…