From Pixels to Actions - Towards Native Vision-Language-Action Policies Paper • 2611.06918 • Published 15 days ago • 60
CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow? Paper • 2610.25078 • Published Oct 18 • 46
OpenMobile-2: Building Versatile Mobile Agents with Scalable Environments and App-Native Tools Paper • 2610.14157 • Published Oct 9 • 44
When History Is Multimodal: Rethinking Context Management for Long-Horizon Agents Paper • 2608.29897 • Published Sep 1 • 1
Glasseek: A Transparent Recipe for Training and Trustworthy Evaluation of DeepSearch Agents Paper • 2610.16998 • Published Oct 12 • 4
Learned Smartphone ISP on Mobile GPUs with Deep Learning, Mobile AI & AIM 2022 Challenge: Report Paper • 2211.03885 • Published Nov 7, 2022
ISE: An Execution-Grounded Recipe for Multi-Turn OS-Agent Trajectories Paper • 2606.11520 • Published Jun 9
ICA: Information-Aware Credit Assignment for Visually Grounded Long-Horizon Information-Seeking Agents Paper • 2602.10863 • Published Feb 11 • 10
Glasseek: A Transparent Recipe for Training and Trustworthy Evaluation of DeepSearch Agents Paper • 2610.16998 • Published Oct 12 • 4