Skip to content

Technical Directions

Robot policies Latest coverage

Robotics coverage relating to robot policies.

776 selected reportsLast 30 days 662 recordsAll coverage 3,792 records

updated

Signal archive — Page 36

2026-10-08Thu701–720 records
  1. arXiv Robotics — research abstracts82

    PhysEvo: Astra Can Act, Let It

    This paper presents PhysEvo, a framework for physical recursive self-improvement (RSI) that allows a frozen model to adapt and improve through task execution and meta-agent feedback. The framework enables joint-level control, evidence-seeking observation, and reusable manipulation skills without model-weight updates or a separately trained action policy. Evaluated on 42 RoboDojo tasks and 8 manipulation tasks, PhysEvo achieves an average score of 68.14/100 and 62.00% success, significantly outperforming existing methods. Deployment on real-world tasks shows 84.00% success across 25 trials.

2026-10-07Wed
  1. Robotics — Robot learning news discovery68

    Our Journey Building Vision-Language-Action Robotics

    This tutorial details the progression from imitation learning to Vision-Language-Action (VLA) models in robotics, focusing on the shift from repeating motions to understanding and executing tasks. It outlines key experiments with models like SmolVLA and π0.5, and discusses challenges such as data quality, model capacity, and simulation-to-real transfer, ultimately aiming to build robots that can see, understand, adapt, and act.

  2. Robotics — Robot learning news discovery72

    Our Journey Building Vision-Language-Action Robotics

    This article details the progression of a robotics team's work from simple imitation learning to developing Vision-Language-Action (VLA) models. It outlines challenges in adapting robotic behavior to new environments, the development of models like SmolVLA and π0.5, and the shift toward simulation-generated data for training. The focus is on building robots that can see, understand, and act autonomously in dynamic settings.

  5. NVIDIA — Robotics model repository updates68

    NVIDIA Releases Cosmos3-Nano Policy Model for LIBERO-10 Environment

    NVIDIA has released a trained policy model for the Cosmos3-Nano framework, specifically adapted for the LIBERO-10 environment. The model was trained on a combination of real and synthetic data, achieving success rates of 93.9% in standard scenes and 84.7% in perturbed scenes. The model is part of a broader study on restyling demonstrations and is available under the NVIDIA Open Model License.

  6. NVIDIA — Robotics model repository updates68

    NVIDIA Releases Cosmos3-Nano Policy Model for LIBERO-10 Sim-to-Real Transfer

    NVIDIA has published a policy model trained on LIBERO-10 data as part of the Cosmos Cookbook, enabling sim-to-real transfer and policy evaluation. The model is exported as a checkpoint with EMA weights and can be loaded by the cosmos-framework policy server. Training data and restyled demonstrations are sourced from the LIBERO_LeRobot_v3 repository.

  7. Chinese robotics — Hardware and sensing68

    2026 Robot Industry 'Brain' Track Review: Two Paths of Unitree and Xian Gong Intelligence

    In 2026, the robot industry shifted from hardware competition to a contest of 'brain' capabilities. The article analyzed the technical stack of 'robot brains,' including perception, cognition, planning, and control layers, and discussed the importance of data loops. The article focused on two companies, Unitree Technology and Xian Gong Intelligence, representing the 'body-centric' and 'brain-centric' approaches respectively, exploring their deployment strategies and data accumulation methods in real-world scenarios.

  8. Tech Xplore — Robotics72

    Threadlike motor uses sliding fibers to drive flexible robotic devices without gears

    Researchers from EPFL's Soft Transducers Laboratory have developed FiberMotor, a threadlike motor that converts electrical energy into motion through the sliding of concentric fibers. This gearless design enables flexible, bidirectional movement, making it suitable for soft robotic applications. The motor's small size and lightweight construction allow for integration into smart textiles and wearable devices, with potential uses in assistive robotics and haptic systems.

  9. Chinese robotics — Deployment discovery68

    Quantopian's 'Physical AI' Real-World Breakthrough: Robot 'Hand-Pressed Lemon Tea' 12-Step Verification

    Quantopian, a Hong Kong-listed company, is advancing its 'physical AI' strategy by testing robotic solutions in real-world restaurant environments. The company's approach emphasizes real-world data and iterative model training to improve robotic dexterity and task execution. A recent demo showed a robot successfully completing a 12-step process of making hand-pressed lemon tea, demonstrating its ability to handle complex, dynamic tasks in real-world settings.

  10. 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.

  11. Robotics — Robot learning news discovery82

    DailyArxiv - Ed1sonChen

    This paper presents a collection of recent research papers and datasets focused on vision-language-action (VLA) models in robotics. Topics include recursive video in-context learning, simulation-guided world models, closed-loop benchmarks for driving, and methods for improving generalization and safety in VLA systems. The work highlights challenges such as task success vs. instruction following, the grounding gap, and the need for efficient and reliable adaptation in real-world scenarios.

2026-10-06Tue
  3. Chinese robotics — Robot learning research75

    Embodied AI Algorithm Stack Overview

    This guide outlines the algorithm stack for embodied AI, from engineering tools and perception to learning and decision-making. It emphasizes the importance of system reliability and data quality, and provides a structured learning path for beginners, including key resources and practical advice.

  4. Chinese robotics — Dataset and collection discovery65

    Daimon Robotics Launches Global First Tactile-Grounded World Model Daimon-TWM

    Daimon Robotics has launched Daimon-TWM, the first global tactile-grounded world model, which integrates tactile perception into perception, decision-making, and action generation. The model is based on the largest tactile multimodal physical world dataset, Daimon-Infinity, and demonstrates significant improvements in physical interaction tasks. The company has also built a global tactile data collection network and established industry benchmarks for physical interaction capabilities.

  5. Robotics — Robot learning news discovery78

    ChunkVLA-AM: Parallel Action Chunking for Vision-Language-Action Models in Additive Manufacturing

    This paper introduces ChunkVLA-AM, a reproducible deployment pipeline for vision-language-action (VLA) models on a FAIRINO FR3 robot in an additive manufacturing (AM) workcell. The system converts real-world demonstrations into OpenVLA-compatible datasets and uses a cloud-edge architecture for inference. It achieves 92.9% success in transferring objects between points in a physical test, with failures attributed to insufficient release-height control. The study also evaluates the model's performance under varying illumination conditions, finding that prediction errors are lowest within a specific luminance range.

2026-10-05Mon
  1. Robotics — Robot learning news discovery78

    Rho: A Foundation for Efficiently Adaptable VLA Models

    Rho is a new foundation model for vision-language-action (VLA) models in robotics, designed to improve adaptation to specific robots and tasks. The paper introduces a training recipe that separates embodiment adaptation from task adaptation, reducing the amount of task-specific data needed for fine-tuning. Rho variants are pre-trained on different robots and show strong performance on physical tasks, outperforming existing models in several benchmarks. The model also supports online adaptation through a latent policy that learns from human corrections.

  2. Robotics — Robot learning news discovery68

    Vision-Language-Action in Robotics: A Survey of Datasets, Benchmarks, and Data Engines

    This paper presents a systematic analysis of Vision-Language-Action (VLA) research, focusing on datasets, benchmarks, and data engines. It identifies key challenges in data fidelity, scalability, and evaluation, and proposes a structured approach to address these issues. The authors release an open-source repository to support the community.

  3. Machine Heart — Robotics on WeChat85

    Can Embodied Intelligence Be Trusted? Sergey Levine Points the Way

    At the 2026 World Robotics Conference, Sergey Levine pointed out that the robotics industry is undergoing a shift from controlling the body to developing decision-making capabilities. He emphasized that robots need to combine prior knowledge with real-world experience to achieve true intelligence. Although robots are still in the foundational technology stage, through data flywheels and scaling, they may eventually enable broader applications.

Timezone · UTC

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