Có lẽ điều giá trị nhất mình mang về sau buổi meetup của @axisrobotics không phải là những tấm ảnh hay những món quà, mà là những người bạn, nhưng người anh em mình đã gặp.
Đến giờ cảm xúc vẫn còn nguyên vẹn và thật sự khó có thể diễn tả bằng lời.
Cảm ơn vì đã đến,
Cảm ơn vì những cái bắt tay,
Cảm ơn những người anh em,
Cảm ơn @axisrobotics ,
Cảm ơn tất cả vì đã làm nên một ngày thật đáng nhớ.
Cảm ơn!
@axisrobotics just made the contributor experience even smoother. 👀
The new “Sign all” feature is live — no more signing approved tasks one by one.
A small update, but a huge quality-of-life improvement for active contributors with hundreds or thousands of trajectories.
Keep shippingggggg
Big congrats to @axisrobotics on the commercial partnership with @boosterobotics
Humanoid hardware meets simulation-powered data infrastructure.
By combining Booster’s robotics stack with Axis’ scalable data engine, this partnership can turn real-robot demonstrations into much larger training datasets for sim-real co-training and better robot policies.
Another strong step toward making Physical AI scalable.
Axis keeps building.
Announcing our commercial partnership with @boosterobotics
Booster builds humanoid robot hardware, OS, and developer tools to make humanoid robots more affordable, reliable, and practical. The partnership centers on using simulation to multiply the value of real-robot data—expanding teleoperation demonstrations into scalable training data across diverse tasks and scenes.
This joint effort powers sim-real co-training and foundation-model development, accelerating progress from hardware iteration to deployable robot policies.
Together, we're building simulation‑powered data infrastructure for Physical AI — making scalable training data accessible to model developers and the broader robotics ecosystem.
@axisrobotics This is exactly the kind of infrastructure Physical AI needs. Quality data compounds, and so does robot intelligence. Excited to see Axis Dataset V1 keep growing 😆
One of the biggest challenges in Physical AI has always been data.
Axis Dataset V1 is an important milestone, showing that large-scale, crowdsourced simulation data can significantly improve robot policies. Seeing π0.5 improve from 83.9% to 88.8% on LIBERO Plus is a strong signal that the data engine is working.
Proud to contribute to the ecosystem and excited to see the dataset continue growing.
Introducing Axis Dataset V1 - the simulation dataset for scalable robot manipulation.
Can noisy, crowdsourced simulation data support embodied pretraining? Yes—with enough coverage and diversity.
Continual pretraining on AXIS dataset V1 lifts π0.5 from 83.9% to 88.8% on LIBERO-Plus, with performance improving consistently as the pretraining data scales from 25% to 100% of the full dataset, showing no clear saturation.
Built with researchers from @GeorgiaTech@UCBerkeley@TAMU@JohnsHopkins@Penn@UMich@NUSingapore@NTUsg.
➡️Related links:
Paper: https://t.co/rbLEnEtPPn
Dataset: https://t.co/Uu7QqPc8xv
Project Page: https://t.co/LwpAopJ4Fh
GitHub Codebase: https://t.co/1napmtKFnA
Putting data quality first has always been my priority when contributing to @axisrobotics.
I’m currently ranked Top 12 on the Points Leaderboard. It may not be the highest position, but almost every point reflects the time, effort and quality data I’ve contributed to the network.
I’m quite satisfied with how the new Point System, launched last night, evaluates contributors and rewards real contributions.
Points will be settled every two weeks, with the next settlement on August 11.
See you in the Top 10 Points Leaderboard and the Top 10 contributor rankings this August.
The Axis Point System is LIVE.
Every valid data contribution is recorded, quantified, and reflected in your Points—so real contributors are recognized for honest, high-quality work that advances robotics.
Check your Points → https://t.co/o543a0DNbW
How it works ↓