Skip to content
#Data and simulation
Oct 24Sat
  1. 简智机器人 · X60

    Gen-HumanEgo tops 550,000 HuggingFace downloads one month after release, ranking second among robotics datasets

    One month after release, GenRobot's open-source Gen-HumanEgo dataset has surpassed 550,000 downloads on HuggingFace, ranking second among robotics datasets. The dataset converts synchronized egocentric video via a Data Foundation Model into a unified token representation aligned across video, motion, hand tracking, depth, and semantic annotations, for robot learning and training. GenRobot says more open datasets will be released later.

Oct 23Fri
  2. 云深处科技 · 微信公众号72

    DEEP Robotics unveils five embodied brain results: world model DR-WM-1.0, VLA model DR-VLA-2.0, and more

    DEEP Robotics has released five research results centered on its self-developed embodied brain, DEEPRobotics-Brain, including the world model DR-WM-1.0, the embodied foundation model DR-VLA-2.0, the embodied agent DR-Agent, the policy optimization framework DR-APO, and the data closed-loop platform DR-Benchmark. The company is also collaborating with seven universities, including MIT, Tsinghua, and Zhejiang University.

    Why it matters: The five results span world models, VLA, policy optimization, and data benchmarks, offering a way to compare the division of labor and metrics across the full embodied brain pipeline.

  3. 亿欧网 · 微信公众号42

    Lingdong General releases iFlyBot-WAM industrial world action model and LindenBot-Harness execution organization system; LDB robots already deployed in factories

    At the 2026 iFlytek 1024 Developer Festival, Lingdong General released the iFlyBot-WAM industrial world action model and the LindenBot-Harness execution organization system, and debuted the LDU environmental data collection solution. The LDB embodied intelligence general-purpose robot has obtained CR certification and EU CE-MD market access, and in B2B warehouse picking has validated an operational path covering autonomous movement between shelves, recognition of goods in different poses, and flexible grasping. Its data collection, training, and inference robot LDT had previously completed validation in small-item feeding scenarios.

  4. 甲子光年 · 微信公众号40

    5th Pazhou Algorithm Competition World Model Ecosystem Partnership Track: Human-centric and Intelligent Driving Tracks Enter Finals

    The World Model Ecosystem Partnership Track, specially established by the 5th Pazhou Algorithm Competition, has entered the finals stage and is scheduled to conclude in mid-November, focusing on two tracks: the human-centric world model and autonomous driving for unmanned mining transport. The human-centric track requires models to infer, from video options, the human behaviors that cause state transitions in household scenes, with a 30-to-1 selection in the preliminary round and a 10-to-1 selection in the finals, and each team is limited to two A800 GPUs for inference; the intelligent driving track uses real multimodal datasets provided by Tage Zhixing to assess detection, segmentation scoring, and safety, efficiency, and path completion.

  5. arXiv cs.RO61

    RGraph-Eval: Evaluating Robot Video Generation with Task Scene Graphs

    The authors propose RGraph-Eval, a reference-demonstration-based framework for evaluating robot videos that uses task scene graphs to explicitly expose interaction errors in generated videos, such as errors involving objects, grasps, and placements. The framework uses a foundation-model pipeline to identify manipulation roles, locate object masks, infer and verify state transitions and action intervals, and reconstruct the 3D motion of manipulated objects. It outputs an interpretable score that covers visible objects, relation changes, and 3D path shape.

  6. arXiv cs.RO74

    LocoManiLab: A unified testbed for data, policies, controllers, and deployment in humanoid loco-manipulation

    LocoManiLab is a unified testbed for humanoid loco-manipulation covering data collection and generation, policy training, evaluation, and real-world deployment. Its task suite includes 234 atomic and compositional tasks across five household scene categories, supported by 51.5 hours of simulated teleoperation data.

  7. arXiv cs.RO60

    STAMP: Humanoid Motion Transitions with Human Motion Priors

    The paper proposes the STAMP framework, which uses a "robot-to-human-to-robot" process to connect independently captured humanoid motions into continuous sequences. Its core is a learned alignment model: after motions are retargeted to a human skeleton, the model ranks candidate positions, headings, and transition durations for the subsequent motion, and a human motion prior then generates the transition under the selected alignment. The study evaluates the tracking success rate of the generated transitions on the Unitree G1 and ROBOTIS K1, and demonstrates long-horizon motion transitions in real-robot demos.

  8. arXiv cs.RO62

    STR: A Spatiotemporal Preference Motion Retargeting Method for Humanoid Loco-Manipulation

    The authors propose a spatiotemporal retargeting method, STR, which uses a vision-language model to select relevant motion signals and body parts and identifies keyframes accordingly, allocating additional solver iterations to keyframes and higher interaction weights to the selected parts to reduce retargeting computational overhead.

  9. arXiv cs.RO67

    FRED: World action model learns future video representations via self-distillation

    The study proposes the FRED framework, which predicts future video representations from noise in a single step via self-distillation, using representations extracted from real future observations as targets, while historical observations provide temporal context for the video model. Across diverse real-world manipulation scenarios, FRED outperforms compared world action models, with notable gains on long-horizon and out-of-domain tasks under zero-shot environment generalization; using only human video demonstrations, it achieves a 50.0% success rate on tasks absent from robot training data, versus 19.4% for the best comparable model.

  10. HuggingFace Daily Papers56

    SkeLAM: A Skeleton-Guided Latent Action Method for Robot World Models

    The paper proposes SkeLAM, a skeleton-guided latent action model for controllable robot world modeling. It decomposes latent actions into a kinematic component supervised by the robot skeleton and a residual component that retains complementary visual dynamics, and uses cross-modal swap reconstruction to make RGB and skeleton kinematic representations interchangeable.

  11. OFweek Robotics62

    FANUC and Hitachi Form Strategic Partnership to Validate Physical AI at Hitachi Factories and Expand to Nine Industries

    FANUC and Hitachi announced a strategic partnership on September 30, under which Hitachi will provide AI technology and FANUC will provide industrial robots to jointly validate and commercially deploy Physical AI. In the validation phase, Hitachi's Ibaraki-area factories will serve as Customer Zero, allowing AI to continuously learn on real production lines the shop-floor rules for grasping parts of different shapes and for setup changes during product changeovers. From fiscal 2027, the partners will jointly expand to global customers in nine industries, including semiconductors, pharmaceuticals, automotive, and shipbuilding.

  12. OFweek Robotics43

    Zhang Yiming personally oversees world model: ByteDance won't build robots, aims to be physical AI's "data nanny"

    ByteDance has chosen to enter the embodied intelligence track with an infrastructure positioning, not building robot bodies but instead laying out physical AI data supply. Zhang Yiming is personally supervising a real-time spatial video generation large model, expected to launch as early as October, used to generate simulated visual data that conforms to physical laws; Seed Robotics, originally led by Li Hang, has been merged into Zhou Chang's management scope, combined with multimodal interaction and world models to achieve a technical closed loop and data reuse.

  13. HuggingFace Daily Papers58

    A Survey Unifying Video Foundation Models Through World Modeling: Reading, Writing, Sharing, and Interacting

    A survey examines video foundation models and video world models through the unifying lens of world modeling, organizing the literature into four complementary operations: reading, which infers world properties from observed video; writing, which synthesizes video from specified or learned conditions; sharing, which integrates understanding and generation through common modeling mechanisms; and interacting, which predicts and updates world states under actions and observations.

  14. 深蓝AI · 微信公众号67

    IROS 2026 NIDAR uses RGB for zero-shot LiDAR point cloud intensity generation

    The IROS 2026 paper NIDAR proposes generating pseudo near-infrared images from RGB images and using intrinsic decomposition to add point-level intensity to existing point clouds; after offline training, it can transfer zero-shot to target scenes. In the Warehouse and Hospital sequences in Isaac Sim, after replacing with NIDAR intensity, the ATE RMSE of ISC-LOAM decreased by about 43.0% and 41.1%, respectively, compared with the simulator's built-in intensity.

  16. Beijing Humanoid Robot Innovation Center38

    Caixin Time X Xiong Youjun: No Bubble Industry, Only Bubble Companies

    In an exclusive interview with Caixin Time, Xiong Youjun, CEO of the Beijing Humanoid Robot Innovation Center, said the humanoid robot industry has no bubble, only bubble companies, and that the share of prototype robots actually doing real work remains limited. He disclosed that about 80% of Tiangong's data training comes from virtual simulation and 20% from real-world scenarios, that the Pelican-Unify unified representation embodied world model can improve execution accuracy but has high computing power requirements, and that the next funding round is already being planned.

  17. HuggingFace Daily Papers75

    DiscoDemo: Using reinforcement learning to generate diverse robot demonstration data to improve imitation learning

    DiscoDemo uses reinforcement learning as a data generator, training a privileged-information latent-conditioned generator that discovers multiple successful solutions through a parallel reverse curriculum and a success-gated diversity objective, then exports them after safety filtering as a static dataset for downstream visuomotor imitation learning. On four manipulation tasks, the method raises downstream success in simulation from 40% for the strongest baseline to 62%, and on a real robot from 39% to 78%. The authors believe that the diversity of successful solutions can improve imitation learning for a pretrained flow-matching VLA, and that it complements corrective data collection.

  18. HuggingFace Daily Papers62

    Ψ-0.6: Adaptive Streams of Thought in a Unified Vision-Language World Model

    The research team proposes Ψ-0.6, a unified vision-language world model that adaptively combines perception, language, geometry, and simulation within a shared reasoning process called streams of thought. These streams alternately generate and interpret images, text, and explicit representations of geometry and motion through a shared token interface. The model surpasses leading open-weight models zero-shot on multi-view spatial reasoning, with further gains from reasoning through geometric intermediate representations.

  19. HuggingFace Daily Papers75

    RynnUnify releases 16B unified embodied model combining visual planning and robot control

    RynnUnify is a 16B unified embodied model that combines embodied understanding with long-horizon visual planning through interleaved video-language reasoning, covering spatial reasoning, physical prediction, and grounding. The authors also built an agentic extension using reinforcement learning and introduced a world-action extension that translates understanding and prediction capabilities into robot control.

  20. 钛媒体20

    Titanium Morning Voice Broadcast: AgiBot Open-Sources First Real-World Physical Agent Dataset; Unitree Technology Clarifies Three Major Layout Directions

    AgiBot's AGIBOT WORLD 2026 fourth installment embodied reasoning dataset has been officially open-sourced, making it the first open-source Physical Agent dataset oriented toward real-world scenarios, with the task focus shifting from manipulation difficulty to decision-making difficulty. Unitree Technology founder Wang Xingxing clarified three major layout directions: improving large-scale mass production capability, advancing the deployment of small physical AI models adapted to its own hardware, and laying out ultra-large-scale computing power clusters, while building a robot self-evolution system of "AI independent R&D—simulation training—real-world testing—feedback iteration."

  21. 钛媒体25

    Titanium Morning Brief: DeepSeek reportedly completes 100 billion yuan financing with CATL and Tencent participating; Foreign Investment Management Measures take effect December 1

    DeepSeek plans to raise about 100 billion yuan (approximately 15 billion US dollars) in its latest funding round, with Tencent and CATL as the largest investors. The deal is expected to be officially completed within the next few days, and the funds will be used for model training, computing infrastructure, and talent recruitment.

  23. Robotics & Automation News72

    Google DeepMind uses Gemini 4 Argon to teach Apptronik's Apollo humanoid new skills from video

    Google DeepMind demonstrated a robot training technique alongside the release of Gemini 4 Argon: after watching raw video of a person playing cornhole, the model analyzes the environment, maps spatial coordinates, and computes full-body joint motion, automatically generating simulation code, then transfers the learned behavior to Apptronik's Apollo humanoid robot, which hit seven consecutive shots in real-machine testing.

  24. Kinetix AI · YouTube62

    KAI Halo case study EgoWild2Dex: training bimanual dexterous manipulation with first-person data

    The KAI Halo case study EgoWild2Dex shows how high-quality bimanual dexterous manipulation capabilities can be derived from highly challenging egocentric data. The project page is https://mmlab.hk/egowild2dex.

    Why it matters: The material presents a complete training path from first-person human data to robotic bimanual manipulation, along with success rates; its transferable aspect lies in its approach to data utilization.

  25. 简智机器人 · X73

    Genrobot open-sources Gen-HumanWholeBody first-person whole-body manipulation dataset

    Genrobot has released and open-sourced Gen-HumanWholeBody, a first-person whole-body manipulation dataset it says has the industry's highest precision, hosted on Hugging Face. The dataset contains 600 hours of whole-body manipulation data, approximately 3 cm whole-body MPJPE, 16,976 real-world scenes, and 1,281 human skills, and is open for research on whole-body motion learning, human-to-robot transfer, and embodied foundation models.

    Why it matters: The original text provides the dataset's scale, precision, and open-access entry point, which readers can use to assess its usability for whole-body motion learning research.

  26. 图灵人工智能 · 微信公众号60

    Jensen Huang GTC Berlin keynote: launches OpenShell, DSX POD, and Agile ONE SN humanoid reference robot

    In his GTC Berlin keynote, Jensen Huang launched several products, including the OpenShell reference architecture that constrains agent permissions outside the sandbox, the DSX POD that can be pre-integrated at the factory and cut deployment time from about six months to as little as 14 days, and the Agile ONE SN humanoid reference robot developed with Agile Robots.

Oct 22Thu
  2. TUM MIRMI62

    TUM RoboGym robotics training center, a collaboration between TU Munich and NEURA, opens

    TUM RoboGym opened in October at the TUM Convergence Center at LabCampus at Munich Airport. It has 48 training units and about 100 robots, and is open to TUM researchers as well as external institutions, universities, and companies.

    Why it matters: The original explains how a large-scale real training facility organizes data collection and sharing against the backdrop of scarce real physical interaction data, and can serve as a reference case for embodied intelligence data infrastructure.

  3. NEURA Robotics74

    NEURA Robotics unveils 4NE1 Gen 4 humanoid robot and opens TUM RoboGym with Technical University of Munich

    NEURA Robotics and the Technical University of Munich (TUM) have opened TUM RoboGym at LabCampus near Munich Airport, covering more than 2,300 square meters. TUM MIRMI is a co-builder, the initial collaboration runs for three years, and the two parties are jointly investing about 17 million euros, of which NEURA is contributing about 11 million euros.

    Why it matters: The original text covers both the new robot's hardware changes and the three sources of training data, letting readers understand NEURA's approach of tying data infrastructure to the hardware launch.

  4. Xbotics具身智能实验室 · 微信公众号77

    How to Choose a Robot Simulator: Same Success, Tenfold Contact Force Gap

    The article interprets a simulator comparison report released by Manda Robotics in early October, running about ten sets of experiments with the same Franka 7-axis robotic arm and the same control program on four physics engines: PhysX, MuJoCo, MuJoCo Warp, and Genesis.

    Why it matters: The article uses controlled comparison experiments under uniform conditions to show that simulator divergence is concentrated in contact forces rather than success or failure, and provides transferable troubleshooting and verification methods.

  5. 机器人前瞻 · 微信公众号58

    Xingyi Intelligence completes tens of millions of yuan in seed and seed+ funding to advance embodied data infrastructure

    Beijing-based embodied intelligence data infrastructure company Xingyi Intelligence announced the completion of consecutive seed and seed+ funding rounds totaling tens of millions of yuan. The seed round was led by Shunxi Fund, and the seed+ round was led by Luster LightTech, with Zhongguancun Capital and First Venture Capital participating. The funds will go toward core algorithm and product R&D, real-world multimodal data infrastructure development, and strengthening customer delivery and commercialization capabilities.

  6. Xbotics具身智能实验室 · 微信公众号62

    HKU team proposes YOCO for dexterous-hand teleoperation mocap calibration with few paired poses

    Researchers from the University of Hong Kong, Yisheng Technology, and Shenzhen Hetao College have proposed YOCO, a few-shot session calibration method for 3D keypoints, aimed at correcting bias in low-cost motion-capture gloves used in dexterous-hand teleoperation. The paper has been accepted to CoRL 2026.

  7. 机器人前瞻 · 微信公众号73

    TeleAI and others propose NEXUS, a cross-terrain perception and motion control foundation, enabling Unitree G1 to perform full-body actions on stairs and slopes

    The China Telecom Artificial Intelligence Research Institute (TeleAI), Gamma Intelligence, Shanghai Jiao Tong University, the University of Hong Kong, and Zhejiang University jointly proposed NEXUS, a cross-terrain perception and motion control foundation for humanoid robots that lets robots autonomously adjust posture and contact based on terrain while preserving motion intent.

  9. 3D视觉工坊 · 微信公众号76

    SenseTime and joint team evaluate GPT-6-Astra embodied capabilities, open-source SenseNova-RoboRSI framework and 27B model

    A joint team including SenseTime and The Hong Kong University of Science and Technology (Guangzhou) released the report "Has GPT-6-Astra Achieved Embodied Intelligence?", systematically evaluating GPT-6-Astra's embodied capabilities across 12 benchmarks, and simultaneously open-sourced the SenseNova-RoboRSI evaluation framework and a 27B embodied model.

    Why it matters: The report breaks model capabilities into testable conditions and provides harness ablation data, so readers can judge the real bottlenecks in embodied capability evaluation and cost.

  10. 机器人产业应用 · 微信公众号62

    Shangpin Home Delivery and Tangyuan Technology Open-Source the WorldSimReady-Home Household Simulation Dataset

    Custom furniture maker Shangpin Home Delivery and simulation data company Tangyuan Technology have jointly released the WorldSimReady-Home dataset for household service robots, freely available on their own platform, HuggingFace, and ModelScope. According to the official description, it contains 100,000 square meters of high-fidelity home scenes, 10,000 interactive object assets, and 1,000 standardized simulation task demos.

  11. 车东西60

    Tesla Q3 net profit falls 43%; Musk says Optimus will be delivered externally next year, with output of 20,000 units per week by end of 2027

    Tesla reported Q3 2026 revenue of $28.046 billion and net profit of $787 million, down 43% year over year. On the earnings call, Musk said Optimus has entered production on a line at the Fremont factory and is expected to be delivered to external customers by the end of next year, reaching an output rate of 20,000 units per week by the end of 2027. He also said an "Optimus Academy" will let the robots observe human tasks and practice repeatedly in simulation and reality to form a reinforcement-learning flywheel, and that the AI 5 chip will be integrated into Optimus around April next year.

  12. 云启资本56

    Tianfu Software Makes Global Debut of Industrial World Model Generator IWM Generator

    At the Industrial World Model session of the 2026 Physical AI and World Models Conference, Tianfu Software made the global debut of its Industrial World Model Generator (IWM Generator), which organizes drawings, engineering models, test data, and operating records into models that can run according to engineering principles.

  13. 钛媒体57

    Chery releases CheryMind, a vertical foundation model for physical AI

    At the 2026 Chery Global Innovation Conference, Chery released CheryMind, a vertical foundation model for physical AI. It serves as the technology foundation of the Yaoguang 2035 strategy, covering automotive R&D and manufacturing and extending to businesses including intelligent driving, intelligent cabins, robotics, eVTOL, and agricultural machinery.