Xiao Xu

A novel reward system boosts Qwen-Image-2.0's performance, achieving a 2.61 point increase in overall quality and significant gains in both text-to-image and image editing tasks.

Bridging the Context Gap in T2I models, Qwen-Image-Agent achieves state-of-the-art performance by intelligently constructing context from user input and external sources.

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

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

Rethinking few-step distillation reveals that the training pipeline's organization is as crucial as the distillation objectives themselves.

Existing text-to-image benchmarks miss the mark on real-world artistic creation, but Qwen-Image-Bench finally provides a creator-centric evaluation that reliably distinguishes state-of-the-art models.

Multi-modal models can now better handle distribution shifts thanks to a new method that explicitly models how different categories are distributed, even when the modalities are asymmetrical.

VLAs can learn to adapt to new environments at test time without any fine-tuning, achieving significant performance gains on robotic manipulation and Atari games.

Representing combinatorial treatments with permutation-invariant embeddings boosts uplift estimation accuracy and stability, especially for rare policies.