Skip to content
Only Research
Sep 23Wed
  1. MagicLab/魔法原子70

    MagicLab analyzes Magic-Brain's in-context visual guidance design; magicVLA grasp success rate rises from 70% to 94.4%

    MagicLab introduces the visual guidance approach in the Magic-Brain framework: a VLM Agent outputs a bbox, and a segmentation model refines it into a three-color mask overlaid on the observation image, converting semantic guidance into visual guidance.

    Why it matters: MagicLab details the design motivation and implementation of Magic-Brain's visual guidance layer, letting readers understand the impact on grasp success rate of converting language instructions into pixel guidance.

Sep 22Tue
  1. Sergey Levine63

    Sergey Levine's team proposes ARLI to guide large robot foundation models with small RL policies

    Sergey Levine shared research conducted with Siemens and Berkeley on how to pair a small RL policy with a large robot foundation model under real-time chunking (RTC). Because the RL policy infers faster, it can observe newer images, thereby steering the foundation model's policy toward better behavior. The introduction he cited notes that asynchronous VLA inference reduces inference latency but breaks the Markov assumption required for reinforcement learning fine-tuning, and ARLI is proposed to address this problem. The project page is https://async-rl-intermediate-information.github.io/ . The work was led by Brian Zhu, Momen Khalil, Emanuele Poggi (Siemens), and @ehharrison4 (Berkeley), among others.

Sep 18Fri
Sep 14Mon
  1. Weekly Robotics45

    Weekly Robotics #376: Humanoid robotics research, FailBench evaluation, and Agility Robotics financial results

    Weekly Robotics issue 376 covers two studies: a dual-depth-camera humanoid from Duke University and a monkey-bar-traversing humanoid from the ETH Robotic Systems Lab. FailBench shows that the best vision-language model for judging robot task success or failure reaches only 0.77 accuracy. Agility Robotics reported 2025 revenue of $1.8M and an operating loss of $140M, with robot-as-a-service pricing at $8,500 per month.

Sep 11Fri
  1. PKU EPIC Lab64

    PKU EPIC Lab proposes UniPart for zero-shot language-driven 3D part segmentation

    PKU EPIC Lab proposes UniPart, a feed-forward cross-modal 3D Transformer conditioned on CLIP text embeddings that uses free-text phrases to select functional parts on point clouds, achieving zero-shot language-driven 3D part segmentation.

    Why it matters: The paper presents a text-driven 3D part segmentation method and a self-built dataset, and validates transfer to real-world language-conditioned grasping.

  2. MIT CSAIL52

    MIT CSAIL releases X-Hinges design software that 3D-prints everyday objects into interactive devices with built-in sensing

    Researchers at MIT CSAIL and Tianjin University have introduced X-Hinges design software, which automatically embeds two different materials when 3D-printing interactive objects, allowing sensor signals and wire-like traces to avoid interfering with each other, and supports up to three axes of motion. The team used it to make a smart glove that transmits grasp-motion data to a humanoid robotic hand, an origami-style desk lamp that senses pinching to adjust brightness, and a detection device that can identify the objects placed on it, and used a computer vision system to align sensor-reading differences among devices from the same batch. The work will be presented at UIST.

  4. PKU EPIC Lab73

    FoldNet++: A Large-Scale Synthetic Dataset for Robotic T-Shirt Folding and Unfolding

    PKU EPIC Lab has released FoldNet++, a large-scale synthetic dataset for robotic T-shirt folding and unfolding, covering 6 robotic embodiments, 1K T-shirts, 1K environmental assets, and 120K richly annotated demonstrations.

    Why it matters: The source gives the dataset scale and the real-world deployment success rate, so readers can assess the feasibility of training garment manipulation policies purely on synthetic data.

Sep 9Wed
  1. MIT CSAIL61

    MIT CSAIL and others launch InstructMesh tool to repair AI-generated 3D models and manufacture them on demand

    MIT CSAIL, together with Google and Northeastern researchers, launched InstructMesh, which combines TRELLIS with GPT-4 to let users repair AI-generated 3D models using natural language before 3D printing them.

    Why it matters: The research presents a new tool and user experiment data, letting readers learn how ordinary users can repair AI-generated 3D models and complete printing.

Sep 7Mon
  1. RLCN强化学习研究 · 微信公众号44

    Roundtable Discussion | RLChina Paper Seminar Episode 168 Live Preview (2026.09.08)

    Episode 168 of the RLChina Paper Seminar will be livestreamed on Bilibili and Tencent Meeting on September 8, 2026, from 16:00 to 16:30, hosted by Hu Zican, a doctoral student at Nanjing University, with master's student Xie Xinyi presenting the paper "Look Before You Leap: Distilling Tree Search into Action Evaluation for Frozen VLA Models."

  2. TUM MIRMI46

    TUM MIRMI uses Embodied Laboratory Intelligence to collaboratively explore the internal mechanisms of cancer cells

    The Cell and BioRobotics Group (CBoRG lab) at the Technical University of Munich (TUM) Munich Institute of Robotics and Machine Intelligence (TUM MIRMI) proposes "Embodied Laboratory Intelligence," in which robotic assistants and virtual agents work together to explore the internal mechanisms of cancer cells.

Sep 5Sat
  1. PKU EPIC Lab68

    GIF: An agentic generation framework for interactive and functional object compositions for robot learning

    PKU EPIC Lab proposes GIF, an agentic generation framework for robot learning that reframes the problem as disentangled reconstruction followed by relative pose recovery. CoGen combines 2D and 3D generative models to produce instance-disentangled meshes and coarse initial poses; GPRM refines the relative poses under joint geometric and physical guidance; and a VLM verifier selects the candidate that best matches the structured specification.

    Why it matters: The paper splits interactive and functional object composition generation into disentangled reconstruction and relative pose recovery, and provides a benchmark of eight contact-geometry classes along with collision-rate figures, which can help readers assess this route to simulation data generation.

Sep 4Fri
  1. 中科慧灵 · 微信公众号65

    CASIA and others propose World4RL: diffusion world model combined with reinforcement learning to optimize robot manipulation policies, real-world success rate rises to 93.3%

    The Institute of Automation of the Chinese Academy of Sciences, Zhongguancun Academy, and the CASBOT team propose World4RL, a policy optimization framework for robot manipulation that combines a diffusion world model with reinforcement learning, enabling robot policies to be optimized end-to-end in imagined trajectories generated by the model without additional real-robot interaction.

    Why it matters: The paper turns the world model from a trajectory-prediction tool used at test time into a reinforcement learning training environment, giving readers a concrete path to understanding how the cost of real-robot interaction can be reduced.

  2. OFweek Robotics72

    Fourier Intelligence, Shanghai Jiao Tong University, and Fudan University Open-Source HandEdit Dexterous Hand Image Editing Dataset and Benchmark

    Fourier Intelligence, together with Shanghai Jiao Tong University, Fudan University, and others, has open-sourced the HandEdit dataset and benchmark, which studies editing first-person human hand images into specified robotic dexterous hand configurations. The data contains more than 200 million image editing samples and over 300,000 video clips, covering 26 URDF configurations, more than 600 scenes, over 1,100 object types, and more than 400 task categories; it uniformly tests 11 image editing methods, sets up two evaluation tracks, Hand-only and Hand-Arm, and combines general similarity, VLM judgment, and embodied task metrics.

Sep 2Wed
  1. PKU EPIC Lab79

    ZETA: A Controlled Study of Zero-Shot Cross-Embodiment VLA Transfer for Tabletop Manipulation

    PKU EPIC Lab published the ZETA study on arXiv (arXiv:2609.02546), proposing a unified definition of zero-shot cross-embodiment VLA transfer and a controlled benchmark covering 14 held-out target embodiments, distinguishing strict zero-shot transfer from pretrain-exposed transfer. Experiments show that local end-effector state-action representations, source embodiment diversity, and auxiliary co-training improve transfer by about 15, 18, and 7 percentage points, respectively, and that adding only 5% target-embodiment data during pretraining raises average target-embodiment progress by 13.4 percentage points.

    Why it matters: The paper provides a unified definition of zero-shot cross-embodiment transfer and a controlled benchmark, allowing readers to understand the actual benefits of different representation and data strategies.

  2. RLCN强化学习研究 · 微信公众号72

    Seven questions on DreamerV3: general reinforcement learning with fixed hyperparameters via a world model

    The RLChina "Seven Questions" column examines DreamerV3, a general reinforcement learning algorithm proposed by Google DeepMind and the University of Toronto. Based on a world model, the algorithm first trains an actor-critic on imagined trajectories, and with a single set of fixed hyperparameters it matches or exceeds domain-specific algorithms across 8 major categories and more than 150 tasks.

Sep 1Tue
  1. Harbin Institute of Technology Robotics73

    HIT's Li Tianlong team publishes research in Science Advances on magnetically controlled CD5+ dendritic cell microrobot immunotherapy

    Li Tianlong's team at Harbin Institute of Technology (HIT) developed an engineered microrobot system based on CD5+ dendritic cells that can achieve precise intravascular navigation through external magnetic field control and activate immune cells in situ; the results were published in Science Advances. Under a conical magnetic field, the system self-assembles into a ribbon-like swarm that migrates downstream; under a rotating magnetic field, it assembles into a chain-like swarm that moves upstream against blood flow; combined with tumor chemotaxis, it achieves targeted enrichment and deep infiltration. In an orthotopic mouse colorectal cancer model, a single injection combined with magnetic navigation reduced the tumor bioluminescence signal by more than 90%. HIT is the first affiliation of the paper.

    Why it matters: The HIT team presents the construction strategy and magnetic navigation results of the CD5+ dendritic cell microrobot, giving readers a look at the specific approach of this therapeutic path of in situ immune activation.

Aug 24Mon
Aug 21Fri
  1. 智源研究院 · 微信公众号75

    BAAI releases FlagEval-Robo embodied intelligence evaluation system and dual leaderboards

    BAAI released the FlagEval-Robo embodied intelligence evaluation system on August 21, launching the FlagEval-Robo-Sim and FlagEval-Robo-Real leaderboards and publishing evaluation results for 12 open-source models. All models underwent equal post-training on the evaluation dataset.

    Why it matters: The evaluation uses a unified dual-track benchmark to quantify the performance gap between simulation and real robots, helping readers judge the limits of how far simulation leaderboard results can be generalized.

Aug 20Thu
  1. 加速进化 · 微信公众号72

    Tsinghua's Zhao Mingguo team publishes Science Robotics paper proposing a unified perception-motion reinforcement learning framework for vision-driven humanoid soccer skills

    On August 19, Professor Zhao Mingguo's team in Tsinghua University's Department of Automation published the paper "Learning Vision-Driven Reactive Soccer Skills for Humanoid Robots" in Science Robotics, selected for the August 2026 humanoid robotics special issue; the robot platform used was a humanoid robot independently developed by Booster.

    Why it matters: The paper breaks down three failure points of the traditional perception-planning-control pipeline and presents a reproducible sim-to-real transfer method.

Aug 19Wed
  1. OFweek Robotics73

    University of Auckland and Xi'an Jiaotong University release Video2Knowledge framework to extract robot task knowledge from monocular phone video

    The LISMS lab at the University of Auckland and Xi'an Jiaotong University jointly proposed the Video2Knowledge framework, which extracts 6D hand and object trajectories from an ordinary monocular RGB video and converts them into robot action sequences. The system uses WiLoR to recover 3D hand meshes, combines OnePoseViaGen for object tracking, and designs a four-stage physics-aware correction module; manipulated-object pose accuracy rises from 52.5% to 59.0% under frontal views.

Aug 14Fri
  1. PKU EPIC Lab66

    PKU EPIC Lab releases HumanTracker humanoid motion tracking benchmark and HumanScore metric

    PKU EPIC Lab proposed the HumanTracker benchmark for evaluating humanoid motion tracking, accepted to ECCV 2026. The benchmark contains approximately 153 hours of optical motion trajectories from multiple professional performers, organized into four motion families with text labels; the accompanying HumanScore metric is trained on 12K motion pairs containing 24K motions, aligns with human preferences, and can reveal contact and stability failures that kinematic metrics often miss, such as foot skating and mistimed landings.

    Why it matters: The original work proposes a benchmark and metric that align humanoid motion tracking evaluation with human perception, and points out that existing kinematic errors miss contact issues such as foot skating.

  2. Harbin Institute of Technology Robotics65

    HIT's Wu Zhiguang team develops adaptive magnetically controlled nanorobot swarm platform, achieving navigation in the spinal canal of mice and non-human primates

    Professor Wu Zhiguang's team at Harbin Institute of Technology (HIT), together with collaborators, developed an adaptive magnetically controlled nanorobot platform that can form swarm motion patterns such as vortex, chain, and ribbon under external magnetic field drive, and achieved targeted navigation within the spinal canal in mice and non-human primates. The results were published in Science Advances.

    Why it matters: The research team released experimental results on navigation of magnetically controlled nanorobot swarms in non-human primates. Readers interested in the medical translation of micro- and nanorobots can learn about its technical approach.

Aug 9Sun
  1. LeadeRobot77

    Mila and Polytechnique Montréal team open-source ~$2,000 robot guide dog Milo, built on Unitree Go2 for autonomous indoor and outdoor navigation

    Mila-Quebec AI Institute and a Polytechnique Montréal team have released Milo, an open-source robot guide dog that they say can cut the roughly $50,000 cost of a guide dog to about $2,000. The hardware base is a Unitree Go2 Air/Pro plus an NVIDIA Jetson Orin Nano; a magnetic-encoder telescoping handle is used to inversely solve the user's 3D position, a bird's-eye-view map is built first and then a navigation policy is run, and the policy is trained in about 10 minutes in GPU-accelerated simulation and deployed zero-shot on the real robot.

    Why it matters: The original article fully breaks down Milo's perception, policy training, and real-world testing details, so readers can reuse its engineering trade-offs in their own navigation projects.

  2. LeadeRobot79

    Italy's IIT and INAIL unveil ergoCub humanoid robot; human-robot collaborative lifting can reduce load on the human lumbosacral joint

    The Italian Institute of Technology (IIT) and INAIL proposed a shared embodied intelligence architecture and built the humanoid robot ergoCub; the research was published in Nature Machine Intelligence.

    Why it matters: The original article lays out methods for building ergonomics into robot hardware optimization along with real-robot data, letting readers understand how design and control are organized around human-robot collaboration.

Aug 8Sat
Jul 27Mon
Jul 15Wed
  1. 正行创新 Striding AI59

    Tsinghua and Striding AI propose self-supervised frame-level advantage modeling algorithm STEAM, achieving a 92.3% success rate on the towel-folding task

    Associate Professor Yu Chao's team at Tsinghua University, together with Striding AI, jointly proposed the self-supervised frame-level advantage modeling algorithm STEAM, which learns frame-level progress signals from the temporal order of expert demonstrations and scores expert data, robot rollouts, and human correction data frame by frame.

Jul 14Tue
  1. Singularity Hub59

    Humanoid robot Surgie performs two live pig gallbladder removals under surgeon teleoperation

    A UC San Diego team completed two live pig gallbladder removal surgeries in an ordinary operating room using the teleoperated humanoid robot Surgie, with surgeons directing and controlling the procedures rather than the robot operating autonomously. Surgie is a modified commercial Unitree G1 that can quickly switch between standard surgical instruments and is smaller in size and footprint than conventional surgical robots such as Da Vinci. Participating surgeons reported intuitive control and reduced physical strain, but also noted overheating and the need for frequent repositioning.

Jul 8Wed
  1. PKU EPIC Lab68

    WAM-TTT: Steering World-Action Models at Test Time by Watching Human Videos

    PKU EPIC Lab proposes WAM-TTT, a test-time training framework that steers world-action models from raw human videos without robot actions, human-side annotations, or task-specific fine-tuning. The method absorbs human videos into a lightweight adaptive memory inside a frozen WAM through self-supervised video prediction, and uses a meta-training stage with paired human-robot data and a key-value memory reconstruction objective to align human demonstrations with robot behaviors. Experiments show it consistently outperforms in-context human-video conditioning baselines across diverse manipulation tasks and generalization settings.

    Why it matters: The original work proposes a method for adapting a frozen world-action model at test time using only unlabeled human videos, without robot demonstrations or fine-tuning, offering methodological reference value for researchers who want to reuse it.

Jul 4Sat
Jul 1Wed
  1. 求之科技 · 微信公众号67

    GeoPredict makes VLA predict future trajectories and 3D geometry before generating actions

    The Chinese University of Hong Kong, Shenzhen and other institutions propose GeoPredict, which explicitly predicts end-effector and joint-level future trajectories before the VLA generates actions, and uses 3D Gaussian modeling for the geometry of the future workspace.

    Why it matters: GeoPredict incorporates future trajectory and 3D geometry prediction into VLA training, giving readers a spatial reasoning approach that does not increase inference overhead.

Jun 29Mon
  1. New Scientist Robots45

    Remote-controlled cockroach swarm can now breathe underwater

    The team of Hirotaka Sato at Nanyang Technological University developed 3D-printed waterproof diving suits for remote-controlled Madagascar hissing cockroaches (Gromphadorhina portentosa), enabling them to walk underwater for up to 3 hours at depths of up to 50 centimeters. The suits supply oxygen through a reaction between hydrogen peroxide and manganese dioxide; five suited cockroaches remained healthy after three days, and reached an underwater speed of 78.4 millimeters per second (87.5 millimeters per second on land). The team next plans to test the approach under extreme space environmental conditions.

Jun 27Sat
Jun 26Fri
Jun 24Wed
  1. Robot.com Holdings Inc.68

    Robot.com releases R-KNav 10,000-hour sidewalk navigation dataset and launches FoMo Challenge

    Robot.com has released the R-KNav dataset and launched the FoMo Challenge, inviting the community to train visual navigation models on real data from its R-kiwi sidewalk delivery robot fleet and to evaluate them in controlled tests on physical robots.

    Why it matters: The data comes from a long-running commercial delivery fleet rather than simulation, and the KPI criteria for physical-robot evaluation have been published, making it possible to assess the reusability of real-street navigation data.

  2. Harbin Institute of Technology Robotics62

    HIT's Zhu Yanhe Team Publishes Paper on Cooperative Reconfiguration for Self-Repair in Modular Robots

    Professor Zhu Yanhe's team at Harbin Institute of Technology (HIT) published a paper in the International Journal of Extreme Manufacturing (IJEM), proposing an autonomous fault-tolerant repair strategy for modular robots based on cooperative reconfiguration, which restores joint fault damage by changing morphology without backup module replacement. The method combines topological description and kinematic self-modeling, particle swarm optimization to search for target configurations, and an improved RRT cooperative reconfiguration algorithm. In experiments, the fault-tolerance upper limit was 7 times higher than that of fixed-configuration redundancy methods. HIT is the sole completing institution, and doctoral student Qi Jian is the first author.

    Why it matters: The paper presents a cooperative reconfiguration fault-tolerance strategy that requires no backup modules and data showing a 7-fold increase in the fault-tolerance upper limit, allowing readers to understand a concrete implementation path for self-repair in robot swarms under extreme environments.

Jun 23Tue
Jun 20Sat
  1. Russ Tedrake59

    Russ Tedrake interprets the Ambient Diffusion Policy paper: the value of suboptimal data varies with noise level

    Adam's new paper proposes Ambient Diffusion Policy, which selectively learns robot policies from suboptimal data through a data-usage approach that depends on noise level. The authors note that when training denoising diffusion models, each data point is used multiple times at different noise levels, and a data point's value may differ by noise level. The method can extract value from a considerable portion of the OXE dataset that was previously discarded, and is complementary to pi0.7's approach of learning conditioned on data quality.