X Square Robot @XSquareRobot

X Square Robot is heading to #RSS2026 Sydney. We’ll be there with two events: 🔔Workshop Sharing At the Diffusion for Robot Learning Workshop, we’ll share our latest work from WALL-OSS-0.5 to WALL-WM, covering VLA foundation models, world action models, and real-world robot generalization. ⏰Time: July 17 morning 🔔X-Harbour Night An intimate RSS tech dinner by Sydney Harbour for researchers, engineers, PhD students, and builders working on robotics, embodied AI, robot learning, VLA, control, and physical-world AI. ⏰Time: July 15, 7:30-10 PM GMT+10 Location: Sydney Harbour Exact venue shared with confirmed guests. RSVP: luma.com/57j5gbnb

[PC] @XSquareRobot is advancing embodied intelligence with high-precision robots designed to perceive, reason and act across home, research, education, logistics and industrial settings. Founded in 2023, it is helping bring general-purpose robotics closer to everyday life. #XSquareRobot #EmbodiedIntelligence #RoboticsInnovation #AIRobotics #Automation #SmartRobots #KTAd #KhaleejTimes Read more in the link below 🔗 khaleejtimes.com/business/tech/…

This year is truly the dawn of the “robot data oil” rush--spawning all kinds of pick-and-shovel tools to dig it up. X Square Robot, which offers in-home humanoid cleaning services, has just launched a full UMI/Egocentric data collection system: QUANXTA Zero. Real robots delivering actual on-site services is one powerful data collection method. Meanwhile, bodiless (no physical robot) UMI/Ego-centric( human-centric)data pipelines are also surging in popularity. Well,everyone’s racing to distill the physical world and extract the data that truly matters for advancing robot intelligence.

X Square Robot @XSquareRobot

a week ago

Introducing the QUANXTA Zero Series, X Square Robot’s next-generation UMI data collection solution for embodied AI. The series includes three products: QUANXTA Zero-G0: VR headset + backpack + dual grippers QUANXTA Zero-G1: headband rig + dual grippers QUANXTA Zero-E0: headband

7 12 137 354K 67

Really glad I finally got to visit @XSquareRobot. We got an early look at XRZero-G0, their open-source full-body data collection and training system that works without requiring a physical robot body. At the time it hadn’t been officially released yet, so we couldn’t share anything haha. Now they’ve officially launched the QUANXTA Zero series — a robot-free embodied data production platform. This is not just a hardware setup. It’s a full-stack system that connects data collection, high-fidelity synchronization, automatic cleaning, intelligent annotation, embodied model training, robot inference, and evaluation loops — all in one pipeline. The goal is very clear: close the “last mile” gap between data and models, and build scalable infrastructure for embodied AI. QUANXTA Zero comes in three setups for different capture scenarios: → G1 (UMI-VIO): dual grippers with a head-mounted rig, balancing quality, usability, and endurance → G0 (UMI-VR): full-body mobile data capture → E0 (Ego): lightweight head-mounted setup for first-person data collection In short, it’s building the data foundation for embodied intelligence. Data is what drives intelligence emergence in large models — both scale and quality matter. If the data flywheel is already spinning, real home robots might be closer than we think. @TheHumanoidHub @dolylupec

We’re excited to announce that we’ve closed our Series C financing at a valuation above RMB 20 billion, with IDG Capital joining this round. This follows previous backing from leading financial and strategic investors including HongShan, Xiaomi, Meituan, Alibaba and ByteDance, marking a strong recognition of our technology and long-term vision for embodied AI.

RoboHub🤖 @XRoboHub

2 weeks ago

X Square Robot just closed its Series C at a valuation above RMB 20 billion, about $2.8 billion 🤖 IDG came into this round. The bigger signal is the cap table. HongShan and Xiaomi were already in across earlier rounds, while Meituan, Alibaba, ByteDance, and Xiaomi have each

2 7 62 13K 20

Thank you for visiting us and for the great conversation. It was a pleasure hosting you and sharing what we’re building. Hope to see you back in China soon — the journey ahead will be exciting.

The Humanoid Hub @TheHumanoidHub

2 weeks ago

That's a wrap on an amazing 17-day journey across China. I didn't do this alone. My friend @dolylupec came along for the whole ride. She's got a background in AI and robotics investing, and I'm grateful she chronicled the tours on her account and pitched in on my channel.

keyboard_arrow_left Previous keyboard_arrow_right Next

24 12 210 73K 37

les noms chinois que vous connaissez, DeepSeek, unitree, agibot…ne sont que la partie visible de l'iceberg, en dessous une nuée de jeunes nés dans les années 90/2000 monte des labos d'ia physique et de hardware dont personne en occident n'a encore entendu parler et croyez moi ça va faire un MASSACRE deux exemples qui donnent le vertige comme ça, x square robot construit un modèle de fondation pour le monde réel, son système (wall a)comprend les relations causales entre les objets, il devine ce qui est caché et corrige ses erreurs tout seul, déjà plus de 400 millions levés avec bytedance derrière (investisseurs 100% chinois) et quasi inconnu chez nous et robotera, sortie des labos de Tsinghua qui fabrique plus de 95% de ses composants en interne et dont la main robotique est si précise que Nvidia et boston dynamics l'utilisent pour leur propre recherche dailleurs j’en ai déjà parlé ici mais le terrain qui les porte est unique au monde, la Chine transforme ses data centers et ses bureaux vides en incubateurs gratuits, puissance de calcul et prêts offerts et surtout elle tient près des 2/3 de la chaîne mondiale des composants de robot, moteurs, réducteurs, batteries terres rares donc un gamin avec un modèle ouvert et cette supply chain peut sortir un robot depuis sa chambre quand il en faudrait des années ailleurs donc le jour où un nom chinois surgit enfin sur votre radar sachez que 10 autres avancent déjà derrière lui, ce que vous voyez n'est que l'écume d'une vague de fond et dans le monde physique elle ne fait que commencer

At X Square Robot, our approach starts from real-world learning. From XRZero-G0’s robot-free + real-robot data framework, to WALL-OSS-0.5’s action tokenization, and WALL-WM’s event-centric world model, we are focused on building robots that can understand physical causality, reason through unfamiliar situations, and generalize across real-world environments. Thanks to Interesting Engineering for featuring X Square Robot in this thoughtful piece.

Interesting Engineering @IntEngineering

3 weeks ago

Two Chinese robotics firms are pursuing radically different paths to zero-shot learning and general-purpose robot intelligence. bit.ly/43QqLxe

0 6 13 4K 0

Love how gentle Xiaoliang from @XSquareRobot treats our dog Bingo while being in action! Living with a humanoid🥰🧹🐕 #futureofhome

.@XSquareRobot has open‑sourced three embodied AI frameworks designed to accelerate real‑world robot learning. Including: 🤖 Wall-OSS-0.5 - Vision-Language-Action model 🤖 WALL-WM - World Action Model 🤖 XRZero-G0 - Robot-free data collection and training framework Its goal is to teach robots to operate in unpredictable, real‑world environments. Learn more: humanoidroboticstechnology.com/articles/from-… #humanoids #humanoidrobot #humanoidrobotics #humanoid

Good job！！！ Our work [XRZero-G0] has been added to this collection.🫡🫡🫡🫡🫡

Arata Horie @ArataHorie

a month ago

Awesome-UMI を公開しました！ 特にヒューマノイドロボットの領域でボトルネックと言われている人間のデモンストレーションデータを収集する装置、「UMI（Universal Manipulation Interface）」系の装置をまとめたウェブサイトです。 まだ完璧ではないベータ版ですが、継続的に追加できればと思います

7 59 439 66K 402

Excited to support the #RSS2026 Diffusion for Robot Learning Workshop! 🚀 Looking forward to seeing more great work on diffusion models for embodied AI and robot learning. Submit your work and join us at RSS!

Kaizhe Hu @hkz222

a month ago

🚀 The #RSS2026 Diffusion for Robot Learning Workshop is now open for submission! …26-diffusion-robot-learning.github.io Diffusion models are the next frontier for embodied AI. Submit by June 24 to win a prize of 1000$ sponsored by @XSquareRobot and join us to hear from our invited speakers!

0 14 78 22K 47

Robot-free demos can be collected through our VR interface, inspected, trained, and evaluated in a closed loop. Cool thing is that a small amount of real-robot data mixed with large-scale robot-free data can reach comparable performance, while reducing real-robot data needs by up to 20x. Code: github.com/X-Square-Robot… Paper: arxiv.org/abs/2604.13001

X Square Robot @XSquareRobot

a month ago

🥳Our third open-source release is here: XRZero-G0. After WALL-OSS-0.5 and WALL-WM, we’re open-sourcing XRZero-G0 to Scale Robot Learning with Interfaces, Data Quality and Ratios XRZero-G0 enables robot-free data collection, trainable policy generation, and real-robot

5 22 81 185K 58

Wall-OSS is now the #1 policy on the zero-shot MolmoSpaces evals. A lot of details in their paper, I recommend checking it out.

X Square Robot @XSquareRobot

2 months ago

We are open-sourcing Wall-OSS-0.5. Pretrain Once, Act Anywhere. Wall-OSS-0.5 is a VLA model for real-world robotic manipulation, exploring whether pretraining alone can produce robot capabilities directly testable on physical hardware before task-specific fine-tuning.

6 25 120 25K 84

🥳Our third open-source release is here: XRZero-G0. After WALL-OSS-0.5 and WALL-WM, we’re open-sourcing XRZero-G0 to Scale Robot Learning with Interfaces, Data Quality and Ratios XRZero-G0 enables robot-free data collection, trainable policy generation, and real-robot evaluation through a closed-loop pipeline: Collection → Inspection → Training → Evaluation Key highlights: 2,000+ hours of validated multimodal demonstrations ~85% effective data yield in controlled settings 10:1 robot-free / real-robot data mixing law Up to 20x reduction in real-robot data needs Zero-shot transfer across robot embodiments Built for scalable, reproducible embodied AI research. Project: x2robot.com/x2go Paper: arxiv.org/abs/2604.13001 Code: github.com/X-Square-Robot… Dataset: 📷huggingface.co/datasets/x-squ… @ComWjm @_akhaliq @HuggingPapers @ModelScope2022 @Xianbao_QIAN @XRoboHub @TheHumanoidHub @chris_j_paxton @IlirAliu_

X Square Robot @XSquareRobot

2 months ago

Introducing WALL-WM, our open-source World Model for embodied AI and the next piece of our open-source robotics stack. Carving World Action Modeling at the Event Joints Read the blog: x2robot.com/en/pages/wm Why it matters WALL-WM shifts robot world modeling from fixed-length

8 40 227 42K 199

🌌 At @saturdayrobotic Saturday Robotics Research Night @CVPR, we hosted Xiaofan Li (World Model Tech Lead @XSquareRobot) for a lightning talk on WALL-WM. TLDR: From next chunk prediction → next event prediction. WALL-WM introduces a new training + inference workflow for world modeling, shifting from rigid frame chunks to semantic event signals. It explores tighter integration of agent intelligence and WAM for improved real-world dynamic perception and prediction. x2robot.com/api/files/file… WALL-WM is a World Action Model (WAM) built on event-level VLA pretraining. Existing WAMs typically: • initialize from multimodal/video foundation models • directly train & infer fixed-length action chunks conditioned on observation + instruction Problem: text, vision, and action lie on different manifolds and temporal scales → direct joint optimization can distort pretrained representations. 💡 Core idea: Event as atomic unit WALL-WM replaces fixed frame-chunk modeling with semantic event modeling. Well-posed event: (c_event, O) → v_event Event enforces: • semantic alignment (language ↔ event meaning) • temporal alignment (vision / action / tactile consistency) → “Carve nature at its joints” (Plato, Phaedrus 265e) 🧠 Architecture • Historical Observations + Executions buffer • Multi-View Video DiT → video latents (world dynamics) • Action Transformer → state-action modeling • Unified Event World Modeling block couples video + action pathways Language stack: • Qwen3.5 + Staircase Decoder in unified embedding space ⚙️ Two inference modes (same event-pretrained backbone) 1. Language-Guided Reasoning (Event Mode) • consumes next-event descriptions • produces variable-length execution chunks • includes explicit temporal event tokens (e.g., “pick…1.6s → fallback → pick…2.4s”) • ON/OFF switch separates reasoning from execution → semantic event rollout 2. Event World Modeling • Video DiT + Action Transformer • purely event-centric rollout of dynamics • no fixed-length chunk assumption in modeling → temporal event rollout 🔁 Key decomposition Semantic path: Language → Event Temporal path: Vision/Action → Event Unified abstraction stack: Pixel → Patch → Frame → Event 🧩 Training philosophy (anti–Bitter Lesson framing) Shift annotation cost → training cost via self-supervised event structure learning End-to-end target pipeline: Reasoning/Grounding → Perception → Future Video → 3D Representation → Action Core principle: “The more we do (preprocessing + structure), the less the model has to infer.” Includes: • normalization • spectrograms • voxelization • tokenization ⚠️ video-only pretraining critique Failure modes: • strong latent distribution assumptions (e.g., SIGReg-style constraints) • semantic rediscovery cost in vision-action alignment • weak coupling between language semantics and temporal execution Examples: VJPEA, LeWorldModel-style approaches Fix: Language acts as semantic tagging over VA event clusters, not temporal supervision signal. 🧬 Representation hierarchy Raw physical signals: vision / audio / action / biological signals ↓ (signal processing + mathematical abstraction) structured modalities ↓ Event layer (highest alignment primitive) 📌 Conclusion WALL-WM is not a chunk-level improvement. It replaces fixed temporal chunking with event-level alignment as the fundamental unit of world modeling. Where prior WAMs learn “what action follows this frame window”, WALL-WM learns “what event is unfolding in the world”. WALL-WM defines the event-based representation primitive for future world models and embodied agents. @CVPRConf #CVPR2026 #WorldModel

keyboard_arrow_left Previous keyboard_arrow_right Next

Junfan Zhu 朱俊帆 ✈️ SIGGRAPH @junfanzhu98

a month ago

🌌 At @saturdayrobotic Saturday Robotics Research Night, we hosted @mli0603 Zhaoshuo Li (Robotics & World Model Tech Lead @NVIDIAAI Cosmos) for a lightning talk on Cosmos 3. Cosmos 3 is a unified omnimodal world model built on a Mixture-of-Transformers (MoT) backbone with

keyboard_arrow_left Previous keyboard_arrow_right Next

1 1 10 2K 4

🥳WALL-WM made it to the alphaXiv Trending Top 10. Instead of predicting every frame uniformly, it learns to focus on the moments that matter. Project: x2robot.com/pages/wm arXiv: arxiv.org/abs/2606.01955 GitHub: github.com/X-Square-Robot… HF: huggingface.co/papers/2606.01… @_akhaliq @HuggingPapers @Xianbao_QIAN @TheHumanoidHub @chris_j_paxton @XRoboHub

X Square Robot @XSquareRobot

2 months ago

Introducing WALL-WM, our open-source World Model for embodied AI and the next piece of our open-source robotics stack. Carving World Action Modeling at the Event Joints Read the blog: x2robot.com/en/pages/wm Why it matters WALL-WM shifts robot world modeling from fixed-length

8 40 227 42K 199

🛬 Landing in Denver for @CVPR! 🤖 Excited to introduce Saturday Robotics Research Night (@saturdayrobotic), co-hosted with @aurorafeng_01 and @ManycoreTech. It might be @CVPRConf's most research-dense side event. We've received 550+ registrations and approved 300+ researchers, engineers, founders, professors. The event has also been organically amplified by both the official @CVPR and @CVPRConf X accounts. Our venue can probably fit only ~100 people comfortably, so if you're coming, please arrive early. We've curated 1 hour of lightning talks packed with frontier research, new model releases, and technical hot takes at the frontier of Physical AI: ⚡ @aurorafeng_01 (Founder @neuralmotion) — NM-GenET Generative video-action model for universal embodiment transfer and cross-embodiment policy learning. ⚡ @mli0603 (Robotics & World Model Tech Lead @NVIDIAAI) — Cosmos 3 A true omnimodal world model unifying language, image, video, audio, and action generation/control in a single architecture. Yes, robot policy control included. ⚡ Xiaofan Li (World Model Tech Lead @XSquareRobot) — WALL-WM Moving beyond next-chunk prediction toward next-event prediction for scalable World Action Models. ⚡ @SourORZ1 (@UMich) Test-Time Scaling for World Action Models via Zero-Shot Geometric Verification. ⚡ @JieWang_ZJUI (@Penn @GRASPlab) Toward a Robotics MMLU: What should evaluation for foundation robot policies actually look like? ⚡ @guocheng_qian (Senior AI Researcher @Snap) Diffusion-DRF: Rich differentiable rewards for video diffusion RL fine-tuning. A rare opportunity to hear what may be coming next in robotics, world models, embodied intelligence, video generation, evaluation, and Physical AI. If you're around the convention center, DM me — happy to grab a ☕ and chat about robotics, world models, embodied AI, or whatever you're building. And if you can't make it, you're always welcome to join our weekly @saturdayrobotic World Models Reading Club in San Francisco. See everyone in Denver. ✈️🔥 #CVPR #CVPR2026 #Robotics #EmbodiedAI #WorldModels #PhysicalAI

keyboard_arrow_left Previous keyboard_arrow_right Next