Tongyi-MAI · 9 records

dateidtitleauthorsvenue↑
2026-10-162610.22200OpenGCUA: Toward a Fully Open Stack for Generalist Computer UseHanzhang Zhou et al.arXiv77
2026-10-162610.20939MAI-Gym: An Open All-in-One Platform for GUI Agent Evaluation and RL Training across Mobile and Computer EnvironmentsPanrong Tong et al.arXiv56
2026-09-252609.29892Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner AgentsTingyu Qu et al.arXiv29
2026-08-252608.23035MobilePA-Bench: Benchmarking Mobile Planner Agents on Complex Real-World TasksYi Zhu et al.arXiv28
2026-08-182608.16887An Empirical Study of Training Pixel-Space Text-to-Image Diffusion ModelsDengyang Jiang et al.arXiv36
2026-06-302606.30084One Forward Beats Two: InnerZoom for Accurate and Efficient GUI GroundingChen Liu et al.arXiv5
2026-06-122606.12575High-Fidelity Two-Step Image Generation via Teacher-Aligned End-to-End DistillationDongyang Liu et al.arXiv12
2026-06-112606.09076Beyond Scalar Rewards by Internalizing Reasoning into Score DistributionsXin Jin et al.arXiv61
2026-05-072605.05204D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion ModelsDengyang Jiang et al.arXiv27