Загрузка видео...

Не удалось загрузить видео

На главную

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...

114,840 просмотров • 3 месяцев назад •via X (Twitter)

Комментарии: 34

Фото профиля Stone Tao
Stone Tao3 месяцев назад

don’t want to put too much shade on the progress (i think sim datasets can be good ideas if executed well!), but improving on libero (which is also a sim benchmark) means very little unfortunately

Фото профиля Thor
Thor3 месяцев назад

Let's goo 🔥

Фото профиля Hazzy
Hazzy3 месяцев назад

you edit too much intern

Фото профиля KABILA
KABILA2 месяцев назад

Cookingg

Фото профиля Jontu
Jontu2 месяцев назад

This is a strong proof that diverse simulation data can meaningfully improve embodied AI.

Фото профиля boyBrands_
boyBrands_3 месяцев назад

He has arrived🔥

Фото профиля Sami
Sami3 месяцев назад

So much excited for this

Фото профиля Apurbo
Apurbo2 месяцев назад

We are so excited about this 🤘

Фото профиля AL Shifat
AL Shifat3 месяцев назад

Axis going to make history

Фото профиля dviet.eth
dviet.eth3 месяцев назад

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 😆

Фото профиля Noodle
Noodle3 месяцев назад

cool to see those teleop sessions actually pushing the libero-plus numbers up

Фото профиля Acee (✱,✱)
Acee (✱,✱)3 месяцев назад

Keep cooking 👌

Фото профиля Piggie 🧡|Bird
Piggie 🧡|Bird3 месяцев назад

Overall success rate increased by 5.8%.

Фото профиля 0xmowra.eth
0xmowra.eth3 месяцев назад

Axis V1 dataset boosting robot learning with crowdsourced simulation data impressive

Фото профиля Red
Red2 месяцев назад

Great work

Фото профиля | MinhDucc | Shark
| MinhDucc | Shark3 месяцев назад

Axis LFG 😍

Фото профиля sasafonda✱,✱)
sasafonda✱,✱)3 месяцев назад

gaxis

Фото профиля namnew
namnew3 месяцев назад

gAxis

Фото профиля VINAY (✱,✱)
VINAY (✱,✱)3 месяцев назад

Axis cooking 🔥

Фото профиля savagezz
savagezz3 месяцев назад

cooking something

Фото профиля Hoangdong (✱,✱)
Hoangdong (✱,✱)3 месяцев назад

gAxis

Фото профиля clover
clover3 месяцев назад

move gaxis

Фото профиля vinhle (✱,✱) 🍚 ⛓ ./
vinhle (✱,✱) 🍚 ⛓ ./3 месяцев назад

Exciting to see the potential of crowdsourced simulation data in robotics, and impressive results from continual pretraining on Axis Dataset V1.

Фото профиля Aurora Lux
Aurora Lux3 месяцев назад

AXIS Dataset V1 is a strong step forward for embodied AI—scalable, diverse simulation data is proving to be a powerful foundation for better robot learning. 🤖🚀

Фото профиля cbeta.base.eth
cbeta.base.eth3 месяцев назад

lets goo team fight fight fight

Фото профиля rara
rara3 месяцев назад

gaxis lfg

Фото профиля Linh (✱,✱)
Linh (✱,✱)3 месяцев назад

Axis Dataset V1 is impressive

Фото профиля Nova
Nova3 месяцев назад

The bot is terrible. Please update it. Its performance is like a primitive human living in a cave 😤

Фото профиля MAS_Enpici (✱,✱)
MAS_Enpici (✱,✱)3 месяцев назад

big update and new insight

Фото профиля yellowopus
yellowopus3 месяцев назад

lets cooking bro

Фото профиля KADAFER
KADAFER3 месяцев назад

let him cook 🔥

Фото профиля dondonat (✱,✱)
dondonat (✱,✱)3 месяцев назад

gaxis keep building

Фото профиля sagar
sagar3 месяцев назад

Let's goo 🙌

Фото профиля Gujil Ruipa
Gujil Ruipa3 месяцев назад

Open datasets like this can accelerate robotics research through collaborative innovation

Похожие видео

In my past research experience, finding or developing an appropriate simulation environment, dataset, and benchmark has always been a challenge. Missing features, limited support, or unexpected bugs often occupied my days and nights. Moreover, current simulation platforms are relatively fragmented—making it challenging to replicate the success of the RT-X dataset in unifying community efforts. Introducing RoboVerse, we provide a unified platform, dataset, and benchmark for scalable and generalizable robot learning. We hope to build a shared foundation to combine the community efforts. RoboVerse includes: MetaSim: We carefully designed a configuration system and a universal interface to align current robotic simulators. With MetaSim, you can use any simulator with the same code—bringing together the community’s diverse efforts under one framework! RoboVerse Dataset and Benchmark: We unify popular simulation environments and benchmarks into a single cohesive system and introduce the RoboVerse dataset—a large-scale, high-quality synthetic dataset. Additionally, we propose a standardized benchmark across both imitation learning and reinforcement learning. A cool feature enabled by our unified framework: Hybrid Simulation! You can now integrate physics engines and renderers from different simulators—e.g., using MuJoCo precise physics with Isaac photorealistic rendering. This not only elevates simulation fidelity but also significantly enhances real-world transfer performance across complex robotic applications. Hopefully, our team’s efforts could serve the robotic community to thrive vibrantly in the years to come. RoboVerse is open-sourced🥳!!! Project Page: Documentation: Github Repo: Paper:

Haoran Geng

84,425 просмотров • 1 год назад