Zhuoyuan Wu

[Zhuoyuan Wu @wuzy2115]

5/5 🧪 Robot policy evaluation.

We deploy OSCAR for real-world robot policy evaluation using the episodes from RoboArena. Our extensive experiments show a high correlation ✅ between virtual assessment in our world model and real-world results, paving the way for a future where robot policies are evaluated entirely within generated environments.

[Zhuoyuan Wu @wuzy2115]

1/5 🚀 Thrilled to open-source OSCAR 🤖 — an action-conditioned world model for robotics. It generalizes across different robot embodiments with precise action controllability. All trained on a single GH200 GPU, and outperforms existing open-sourced baselines, which have larger model capacity and need more compute.

Everything is public:
📄 Paper: https://t.co/IQkeWVR2XL
🌐 Project: https://t.co/rbJJiSQfPu
💻 Code: https://t.co/cpp7zxNwqv
🤗 Robot data: https://t.co/fMUZ9Ic59v
🤗 Human data: https://t.co/ouFbu9KOJl
🤗 Weights: https://t.co/zIqTtySxtX

#Robotics #WorldModels #AI #OpenSource

[Zhuoyuan Wu @wuzy2115]

4/5 📈 Results.

With our cleaned data and kinematic skeleton conditioning, we finetune the Cosmos2.5-2B model on a single GH200 GPU ⚡. Our method delivers superior quality on action following, appearance, and motion consistency, pushing SOTA performance with a fraction of the compute. 🏆