Zhuoyuan Wu

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.

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. 🏆

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: arxiv.org/abs/2606.04463 🌐 Project: wuzy2115.github.io/oscar-project-… 💻 Code: github.com/wuzy2115/oscar… 🤗 Robot data: huggingface.co/datasets/zywu2… 🤗 Human data: huggingface.co/datasets/zywu2… 🤗 Weights: huggingface.co/zywu2115/OSCAR… #Robotics #WorldModels #AI #OpenSource