Search papers, labs, and topics across Lattice.
Fudan University
2
0
5
17
Even the best vision-language models struggle with reliable evaluation of computer-using agents, but OS-Shepherd models offer a low-cost solution that matches their performance.
PPO's fixed clipping hurts exploration by squashing high-reward, low-probability actions, but BandPO fixes this with probability-aware bounds that boost performance.
