Skip to content

Technical Directions

VLA Latest coverage

Robotics coverage relating to vla.

414 selected reportsLast 30 days 385 recordsAll coverage 832 records

updated

Signal archive — Page 21

2026-10-07Wed401–414 records
  1. Robotics — Robot learning news discovery82

    Perturbot: Breaking Shortcut Priors in Vision-Language-Action Policies

    This paper identifies modality shortcuts in vision-language-action (VLA) policies, where policies rely on superficial cues like visual, lexical, or motor signals rather than task-relevant evidence. The authors propose Perturbot, a method that introduces task-preserving wrist-view perturbations, enriched instructions with decision-relevant captions, and relabeled trajectory segments to reduce reliance on these shortcuts. They also introduce GroundFscore, an offline score to evaluate how much a policy depends on shortcuts. The work provides a framework for training and evaluating VLA policies that disentangle decisions from shortcut priors while preserving responsiveness to task-relevant evidence.

2026-10-05Mon
  1. Robotics — Robot learning news discovery68

    A Curated List of Vision-Language-Action (VLA) Models for Robotics and Embodied AI

    This GitHub repository is a curated list of Vision-Language-Action (VLA) models, architectures, benchmarks, and datasets for robotics and embodied AI, with a focus on 2025–2026. It includes foundational models like RT-1, RT-2, and OpenVLA, as well as advanced models such as π0, GR00T N1, and Gemini Robotics. The repository also features architectural innovations, benchmarks like LIBERO and RoboTwin 2.0, and open-source code for many of the models.

  2. NVIDIA — Robotics model repository updates75

    NVIDIA releases G1-Factory-GR00T-1.7 policy for Unitree G1 humanoid

    NVIDIA released G1-Factory-GR00T-1.7, a GR00T N1.7 vision-language-action policy fine-tuned on nine simulation tasks for the Unitree G1 humanoid robot. The model takes head-camera images and joint states to predict whole-body actions and serves as a baseline for reinforcement learning post-training. It is available under the NVIDIA Open Model License for commercial and non-commercial use, with evaluation results reported only in simulation.

2026-09-22Tue
  1. NVIDIA — Robotics72

    NVIDIA Isaac ROS 5.0 Advances Agentic, Open Source Robotics Development

    NVIDIA Isaac ROS 5.0, released at ROSCon 2026, enhances open source robotics development with agentic workflows, GPU acceleration, and new skills for perception and manipulation. It supports ROS Lyrical and Ubuntu 24.04, enabling developers to build and deploy high-performance robotics applications using NVIDIA's accelerated computing and AI models.

2026-07-28Tue
2026-07-07Tue
2026-03-09Mon
  1. NVIDIA — Robotics model repository updates75

    NVIDIA releases SOMA-X v0.3.0 unified parametric human body model framework

    NVIDIA has released SOMA-X v0.3.0, a unified framework for parametric human body models that decouples identity representation from pose parameterization. The release supports six identity backends including SOMA-shape, SMPL, and SMPL-X, and introduces native hand layers with interoperability for user-supplied MANO models. Designed for commercial use under the Apache 2.0 license, the model enables real-time simulation pipelines on NVIDIA GPUs with throughput exceeding 7,000 meshes per second on an A100. The core pipeline relies on closed-form analytical modules rather than learned neural networks, though an optional shallow MLP can be enabled for surface correction.

2026-03-05Thu
  1. Hugging Face — Robotics68

    Bringing Robotics AI to Embedded Platforms: Dataset Recording, VLA Fine-Tuning, and On-Device Optimizations

    This tutorial explores the challenges of deploying VLA models on embedded robotic systems, including dataset recording best practices, fine-tuning techniques for ACT and SmolVLA, and real-time performance optimization using the NXP i.MX 95 SoC. It emphasizes asynchronous inference and hardware-aware scheduling to improve control and reduce latency.

2026-01-09Fri
  1. LeRobot — Model repository updates78

    LeRobot - Model Repository Updates

    LeRobot introduces a Vision-Language-Action (VLA) policy called PI0Fast, which predicts continuous robot actions via autoregressive next-token prediction over FAST action tokens. The repository includes training and inference code, along with instructions for evaluation in simulation and real-world settings.

2025-10-01Wed
  1. LeRobot — Model repository updates75

    LeRobot: π₀.₅ VLA Model Release for Long-Horizon Tasks

    LeRobot has released a Vision-Language-Action (VLA) model, π₀.₅, trained on robot demonstrations and multimodal data for long-horizon tasks. The model supports flow-matching for action prediction and is compatible with the original reference code. It is intended for fine-tuning on specific use cases and includes both inference and training examples.

2025-09-09Tue
  1. LeRobot — Model repository updates75

    LeRobot Releases π₀.₅ Vision-Language-Action Model for Physical Intelligence Tasks

    LeRobot has released π₀.₅, a Vision-Language-Action (VLA) model with open-world generalization capabilities, trained on robot demonstrations and multimodal data. The model supports flow-matching for action prediction and is intended as a base model for fine-tuning. Code and documentation are provided for inference and training.

  3. LeRobot — Model repository updates78

    LeRobot Releases π₀.₅ (Pi05) Model for Vision-Language-Action Tasks

    LeRobot has released the π₀.₅ (Pi05) model, a Vision-Language-Action (VLA) model with open-world generalization capabilities. It is trained on robot demonstrations and multimodal data to execute long-horizon tasks in real-world environments. The model supports flow-matching for action prediction and is intended as a base for fine-tuning on specific applications. The implementation is compatible with the original reference code and includes instructions for inference and training.

  4. LeRobot — Model repository updates65

    LeRobot/pi0_base: Vision-Language-Action Foundation Model for Robot Control

    LeRobot releases the π₀ model, a Vision-Language-Action (VLA) foundation model that jointly reasons over vision, language, and actions to control robots. It serves as the base architecture for π₀.₅'s open-world generalization. The model supports training, inference, and real-world evaluation with detailed code examples and setup instructions.

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).