Skip to content
#Robot models and learning
Jul 30Thu
  1. Apptronik Social57

    Apptronik Congratulates Google DeepMind on Gemini Robotics 2 Launch; Apollo 2 Demonstrates Whole-Body Reasoning and Dexterous Manipulation

    Google DeepMind has released Gemini Robotics 2, and robotics company Apptronik shared its congratulations. The quoted content describes it as a new generation of physical AI that brings whole-body intelligence to humanoid robots, with advanced dexterous manipulation and multi-robot collaboration capabilities. Apptronik said that seeing Apollo 2 use advanced reasoning to complete whole-body tasks and achieve fine dexterous manipulation is a huge leap.

Jul 28Tue
  1. 北京智在无界科技有限公司64

    BeingBeyond releases Being-H0.8, the first implicit world action model to introduce touch in pretraining

    BeingBeyond has released Being-H0.8, which it calls the world's first implicit tactile world action model, unifying vision, touch, action, and future state changes into a single latent space, with pretraining on more than 500,000 hours of first-person human video.

    Why it matters: The model brings touch into large-scale pretraining along with a self-developed pseudo-labeling system, letting readers understand the path by which tactile data goes from a lab signal to a scalable pretraining resource.

Jul 27Mon
  1. 线性资本43

    Zhengqi Future Completes Several Hundred Million Yuan Angel Series Financing, Linear Capital Invests in Two Consecutive Rounds

    Physical AI company Zhengqi Future completed several hundred million yuan in angel series financing within 8 months, with Linear Capital investing in two consecutive rounds; core investors also include CDH VGC Investment, SAIC Group Hengxu Capital, and Chow Tai Fook Investment. Its self-developed QUORRA DoorMind world model and QUORRA FLAT motion execution platform have entered testing in multiple scenarios, and its first short-distance mobility robot, QUORRA X5, has secured commercial orders in more than ten overseas countries and regions. The funds will be used for model and platform R&D, QUORRA mass-production manufacturing, and global online and offline channel deployment.

Jul 24Fri
  1. 云启资本42

    Yunqi Quarterly Q2 | Midsummer, the future takes on more concrete forms

    Yunqi Capital released its Q2 quarterly report, disclosing progress by its portfolio companies in three areas: financing, products, and deployment: the post-investment valuations of X Square Robot and Zhineng Pingfang Technology both exceed 20 billion yuan, Stardust Intelligence completed a Series B round of over 1 billion yuan, and Nova Fusion has raised a cumulative 1.2 billion yuan.

  2. 具微科技52

    Juwei Technology releases industrial embodied brain MicVLC

    Juwei Technology has released MicVLC, an industrial embodied brain, and proposed for the first time the VLC (Vision-Language-Calling) architecture paradigm, which decouples the cognitive layer from the execution layer, in contrast to traditional VLA models that deeply couple Action with the low-level control of specific terminal actuators.

  3. Fudan Multimodal Embodied Intelligence30

    Fudan Institute of Intelligent Robotics and Advanced Manufacturing Innovation showcases multiple achievements at 2026 World AI Conference

    During the 2026 World AI Conference, Fudan University's Institute of Intelligent Robotics and Advanced Manufacturing Innovation presented achievements in embodied intelligence, physical intelligence, and related areas. Professor Su Hao delivered a keynote speech on physical intelligence at the main forum; Sudo Technology made its first public appearance in China and demonstrated a four-robot collaborative battery module assembly production line; and Zhang Lihua's team's intelligent inspection robot dog appeared on a CCTV program.

Jul 22Wed
  1. OFweek Robotics68

    Six guests at WAIC Embodied AI Forum predict robots' ChatGPT moment is still two to five years away

    At the 2026 WAIC Embodied AI Forum, six guests predicted that robots' ChatGPT moment is two to five years away, with Yao Maoqing saying two years and Xu Danfei saying five, though no consensus has formed on the criteria for judging it. The discussion focused on three main threads: the data flywheel must be built through real deployments rather than data collection alone; VLA and World Action Models (WAM) are not opposites, with PI continuing the VLA path through π0.7 while Dyna turns to WAM; full-stack capability is becoming a trend, but a full-stack business model is not necessarily the only answer.

  2. 云启资本58

    Yunqi Attention WAIC World Model Special Recap: Three Talks and a Roundtable Discussion

    Yunqi Capital, together with NICE Academic, held a World Model BBQ event during WAIC, receiving over 500 registrations and drawing more than 100 attendees. Chen Siheng of Shanghai Jiao Tong University introduced the paths of EmboMatrix, an embodied decision-making training ground, and EmboCoach, a tool for batch training of VLAs.

  3. Holiday Robotics62

    Holiday Robotics's FlashSAC Paper Wins RSS 2026 Best Paper Award, a First for a Korean Team

    A paper by Holiday Robotics researchers won the Best Paper Award at RSS 2026, a first for a Korean team. Its FlashSAC is a reinforcement learning algorithm that shortens humanoid robot walking training from several hours to a few minutes, overcoming the trade-off between learning speed and stability.

    Why it matters: The original text provides verifiable details such as FlashSAC's training speed and validation scope, letting readers understand how reinforcement learning algorithms can shorten the humanoid robot policy development cycle.

Jul 20Mon
  1. 石麻笔记 · 微信公众号74

    Interview with Dai Ruoli: ModalityNet compiles human data into learnable robot data

    Shima Notes conducted an in-person interview with Noitom founder Dai Ruoli, interpreting the ModalityNet dataset it has launched. The dataset contains three types of data that can be reused across embodiments: MOV large-space interaction data (about 8,000 hours, compared with about 70-plus hours for AMASS), OM dexterous manipulation data (sub-millimeter hand precision, including tactile and spatial audio), and ITW in-the-wild subjective vision data; the public portion is about 1/20 of the amount actually held.

Jul 19Sun
  1. 极智嘉47

    Geek+ unveils Gravity embodied AI framework and "One Core, Dual Engine" strategy at WAIC 2026

    Geek+ unveiled Gravity, its unified embodied AI framework, and Gravity 4D, its first core foundation model, at the 2026 World Artificial Intelligence Conference (WAIC), targeting long-horizon, complex physical tasks. Gravity 4D represents a fundamental leap from predicting what the world looks like to understanding how the physical world behaves, and the company also proposed its "One Core, Dual Engine" strategy.

  2. 它石智航 · 微信公众号56

    TARS Robotics wins WAIC SAIL Star at AWE, A1 robot demonstrates wiring harness assembly and dexterous-hand magic

    TARS Robotics' general embodied large model AWE won the SAIL Star at WAIC 2026, becoming the only embodied intelligence foundation model to receive a SAIL Award this year. At the booth, the A1 robot used AWE to complete tasks such as packing phones into boxes, packing backpacks, and sorting small screws, and replicated a wiring harness factory's circular assembly line 1:1, with multiple A1 robots collaborating under AWE 3.5 to complete flexible wiring harness assembly.

Jul 17Fri
  3. Sunday Robotics78

    Sunday Robotics previews ACT-2 model with 99.1% zero-shot success rate in clothes folding

    Sunday Robotics previewed its robot model ACT-2, which achieved a 99.1% (±0.3%) zero-shot success rate in clothes folding across diverse unseen home environments with zero per-home adaptation, covering 9 clothing categories across 785 autonomous attempts, with an average folding quality of 4.72/5.

    Why it matters: The official disclosure provides detailed quantitative data on pretraining scale and generalization gaps, allowing readers to understand how a small amount of demonstration data transfers to unseen environments.

Jul 15Wed
  1. RoboParty · 微信公众号63

    RoboParty completes nearly 500 million yuan in funding to advance open-source humanoid robot platform and HFM R&D

    RoboParty recently completed consecutive Angel++ and Pre-A funding rounds totaling nearly 500 million yuan. The Angel++ round was joined by existing shareholders including Shunwei Capital, Matrix Partners China, Xiaomi Strategic Investment, and SenseTime Guoxiang, while the Pre-A round was exclusively invested by CATL. Lighthouse Capital served as incubator and financial advisor.

  2. X Square Robot/自变量机器人70

    X Square Robot releases DMuon optimizer, improving embodied model distributed training efficiency by about 30%

    X Square Robot has released the DMuon optimizer, a distributed adaptation of the Muon optimizer that improves embodied model distributed training efficiency by about 30%, reducing per-step training time from 2.2 times that of AdamW to 1.02 times. The improvements include Owner-centric execution and communication optimization, automatic selection of compute kernels based on matrix shape, and balanced workload distribution across Owners. DMuon is fully open source and compatible with Muon usage, requiring only 3 lines of code to introduce into training.

    Why it matters: The original article explains the sources of Muon's overhead in distributed training and three improvement approaches, which embodied model teams can use to assess the feasibility of improving training efficiency.

  3. 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.

  4. LimX Dynamics61

    LimX Dynamics releases humanoid VLA model V³-0 alongside COSA 0.5

    LimX Dynamics released V³-0 alongside COSA 0.5, a humanoid VLA technology stack in its VLA-tools layer that supports mobile whole-body manipulation, running on the 43-DOF LimX humanoid robot Oli.

    Why it matters: The original text explains the architectural composition of V³-0 and its human-in-the-loop real-robot reinforcement learning loop, which lets readers understand where the mobile whole-body manipulation capability comes from.

Jul 14Tue
  1. PsiBot/灵初智能81

    Lingchu Intelligence and PKU-Lingchu Joint Lab Release and Open-Source Dual Dexterous Hand General Manipulation Model and Full-Stack System EgoSteer

    Lingchu Intelligence and the PKU-Lingchu Joint Lab released and open-sourced the dual dexterous hand general manipulation model and full-stack system EgoSteer, comprising the data processing pipeline EgoSmith, the robotics technology stack Robot-Stack, and a world-model-enhanced VLA model.

    Why it matters: The original text presents a complete chain from human-hand data processing and real-robot correction to model architecture and experimental figures, allowing readers to assess its reproducibility value.

Jul 13Mon
Jul 12Sun
  2. 蚂蚁灵波科技 · YouTube67

    LingBot-VA 2.0 released: a world-action model for physical-world robot control

    Ant Group's Lingbo Technology has released LingBot-VA 2.0, which it describes as a world-action model pretrained from scratch for physical-world robot control, running real-time inference at 150 Hz on a single GPU and able to play air hockey, tidy desks, and pick chips from a conveyor belt.

    Why it matters: It provides a same-benchmark success-rate comparison with π0.5 and single-GPU real-time inference metrics, helping to gauge where this kind of world-action model stands in terms of performance.

Jul 9Thu
  1. 蚂蚁灵波科技 · YouTube70

    Ant Lingbo Open-Sources Embodied Video Foundation Model LingBot-Video, Scoring 0.620 on RBench

    Ant Lingbo has open-sourced LingBot-Video, a MoE video foundation model for embodied intelligence. It uses a DiT plus sparse MoE architecture with 30B parameters and 3B activated at inference, and is claimed to be about 3 times more efficient than dense models.

    Why it matters: The material provides comparison scores on RBench and the scope of the open-source release, helping readers judge the starting point and usability of this embodied video model.

Jul 8Wed
  1. Hugging Face LeRobot82

    LeRobot v0.6.0 released with multiple breaking changes and new policies

    Hugging Face LeRobot has released v0.6.0: pip install lerobot no longer includes dataset and training dependencies by default, requiring extras to be added as needed; GR00T N1.5 has been replaced by N1.7; and the minimum PyTorch version has been raised to 2.7.

    Why it matters: The release notes list specific breaking changes and migration methods, so developers in the middle of upgrading can use them to check their installation, configuration, and training workflows.

  2. 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.

  3. 蚂蚁灵波科技 · YouTube69

    LingBot-VLA 2.0 open-sourced: supports 20 robot configurations and full-body control

    LingBot-VLA 2.0 has been open-sourced. It is a next-generation embodied foundation model pretrained on 60,000 hours of real-world physical data, built on a redesigned data processing pipeline curated from 90,000 hours of robotic data and 20,000 hours of egocentric human videos.

    Why it matters: The public weights and GM-100 results let readers directly compare its gap with π0.5 and GR00T N1.7 on long-horizon mobile manipulation.

Jul 7Tue
  1. 智源研究院 · 微信公众号74

    BAAI releases multimodal representation world model Wujie·RoboBrain Orca

    BAAI has released the multimodal representation world model Wujie·RoboBrain Orca. Its core idea is to shift from predicting the next word, next frame, or next action to predicting the next world state: it first learns a unified latent world representation, then adapts to downstream tasks through three lightweight readout modules for text, images, and actions.

    Why it matters: The original text explains the training paradigm shift from predicting the next output to predicting the next state, as well as the approach of a frozen backbone plus lightweight readout modules, which is transferable to readers following the world model route.

Jul 6Mon
  1. X Square Robot/自变量机器人67

    X Square Robot completes four financing rounds, valuation exceeds 20 billion yuan

    Shenzhen embodied intelligence company X Square Robot has completed four consecutive financing rounds from Series B through Series C, all fully closed, with a valuation exceeding 20 billion yuan, making it the first embodied intelligence company in the Guangdong-Hong Kong-Macao Greater Bay Area to surpass a 20 billion yuan valuation with the rounds closed. There are more than 30 investors, including the National Artificial Intelligence Industry Investment Fund, Sequoia China, and Xiaomi Strategic Investment. The company released its embodied large model WALL-B in April 2026, partnered with 58 Daojia to provide robot home housekeeping services to nearly 1,000 households in Shenzhen, and partnered with SF Express on flexible intelligent sorting and parcel feeding.

    Why it matters: The report details how all four financing rounds were fully closed and the deployment scenarios, giving readers a sense of how substantive the company's financing progress really is.

  3. 蚂蚁灵波科技 · YouTube32

    Ant Lingbo Technology LingBot-VA: The First Causal Video-Action World Model for Generalist Robot Control

    Ant Lingbo Technology has launched LingBot-VA, an autoregressive framework that unifies video world modeling with policy learning, jointly learning future frame prediction and action execution. Evaluation results show that the framework demonstrates clear potential in long-horizon manipulation, data-efficient post-training, and generalization to new scene configurations.

Jul 4Sat
Jul 3Fri
  1. 求之科技 · 微信公众号12

    Qiuzhi Technology embodied intelligence VLA bootcamp recruiting: 4 days from simulation to real-robot deployment

    Qiuzhi Technology's ROS summer school "Embodied Intelligence VLA Bootcamp" is now recruiting. The course runs July 20 to 23, 2026 at Zhejiang Sci-Tech University, lasts 4 days, is free of charge, and participants cover their own food and lodging. The course covers vision-language-action (VLA) large models, high-fidelity simulators, robot kinematics modeling, MoveIt2 motion planning, and the Linux/ROS 2 development environment, guiding students through hands-on practice of the full pipeline from perception and decision-making to execution, as well as calibration, control adaptation, and error correction in Sim2Real transfer.

Jul 2Thu
  1. X Square Robot/自变量机器人67

    X Square Robot releases cross-modal embodied action tokenizer X-Tokenizer, with 13.5% gain in multimodal alignment and 8.25% gain in long-horizon tasks

    X Square Robot has released X-Tokenizer, a cross-modal embodied action tokenizer built on a semantic residual quantization (SRQ) architecture and pretrained on 17 robotic arm series and 2.4 million trajectories (including 2 billion action frames). Compared with FAST, it improves multimodal alignment by 13.5% and long-horizon task performance by 8.25%, and scores 82.8 on RoboTwin 2.0; in real-robot tests it outperforms comparison methods such as FAST and 4-level RVQ.

    Why it matters: The original text presents X-Tokenizer's method design and comparative experimental data, letting readers understand the action tokenizer's multimodal alignment approach and its reusable effects.

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.

  2. YOUIBOT/优艾智合53

    Youibot Releases FabriX Industrial Embodied Intelligence Foundation Model and Edge Industrial Humanoid Robot

    On June 29, Youibot released the industrial embodied intelligence foundation model FabriX and the industrial humanoid robot Edge. FabriX uses an edge-cloud hybrid architecture, combining a multimodal MoE foundation model with deterministic control, and proposes an ecosystem partner plan to empower 10,000 industrial sites over three years.

Jun 30Tue
  1. Figure AI63

    Figure 03 arrives at BMW Spartanburg plant for sequencing sorting tasks

    Figure AI announced that the Figure 03 humanoid robot has entered Hall 52 at BMW Group Plant Spartanburg, shifting from the sheet metal loading and unloading work previously done by Figure 02 to a sequencing sorting use case; Figure 02 took part in assembling 30,000 vehicles last year.

    Why it matters: The company has released details of Figure 03's deployment scenario at the BMW plant and of its Helix 02 drive, so readers can learn about progress in bringing humanoid robots from material handling to complex sorting.

Jun 29Mon
Jun 27Sat
Jun 26Fri
  1. OFweek Robotics59

    Qualcomm explains robot System 0, 1, 2 architecture, releases Dragonwing IQ10 reference design

    At its Investor Day, Qualcomm laid out its robotics strategy, splitting robot computing architecture into a System 2 reasoning layer (18-core CPU, 700 TOPS multi-NPU, 64GB memory, over 270GB/s bandwidth, running VLA large models), a System 1 action layer, and a System 0 execution layer, built on deterministic Ethernet with nanosecond-level synchronization and multi-sensor ASICs.