Jiazhao Zhang

Achieving 89.4% success in single-view embodied visual tracking, ReferTrack outperforms multi-camera systems by grounding tracking decisions in real-time visual data.

Instantaneous grasp trajectory predictions enable robots to adeptly handle dynamic objects, achieving unprecedented performance in mobile manipulation tasks.

PAIWorld resolves critical multi-view inconsistencies in robotic manipulation, outperforming existing models and unlocking new applications in the process.

Qwen-RobotNav redefines navigation by allowing real-time reconfiguration of strategies, achieving unprecedented flexibility and performance across diverse tasks.

Qwen-RobotManip achieves a 20% relative improvement over the previous state-of-the-art in robotic manipulation, showcasing unprecedented generalization capabilities from diverse, open-source datasets.

Language-driven video generation in Qwen-RobotWorld achieves unprecedented accuracy in predicting robotic actions, outperforming existing models across key benchmarks.

AllDayNav achieves nearly 100% success in lifelong navigation by using a self-evolving memory system that outperforms traditional mapping techniques.

One model to control them all: Qwen-VLA achieves impressive zero-shot generalization across diverse robotic tasks and embodiments by unifying vision-language-action modeling.

Photorealistic simulation with Gaussian Splatting and drivable avatars closes the reality gap, enabling embodied agents to learn human-aware navigation policies that generalize better to the real world.

Training robots in a photorealistic Gaussian Splatting simulator transfers surprisingly well to the real world, boosting scene understanding and navigation performance.

A single spatial token, learned via occupancy prediction on a massive dataset, is surprisingly effective at injecting crucial spatial awareness into vision-language navigation, leading to state-of-the-art performance.