Long Li

Frontier generalist computer-use agents, previously closed-source, are now open for systematic study and community development, with the first release achieving significant performance gains.

Training GUI agents in mocked environments yields substantial real-world performance gains, bridging the gap between simulation and physical device interaction.

Current omnimodal video embeddings struggle with physical reasoning, yet specific physics can be recovered and leveraged to significantly improve the physical fidelity of generated videos.

Categorical value learning can significantly enhance the performance of PPO critics in reinforcement learning, leading to better calibration and lower variance in advantage estimation.

Qwen-UI-Agent outperforms leading models in mobile and cross-platform tasks, achieving up to 97.5% accuracy on key benchmarks.

PPO can be made sample-efficient and stable for long-horizon reasoning in LLMs by treating the problem as a sequence-level contextual bandit, sidestepping the need for computationally expensive multi-sampling.