Search papers, labs, and topics across Lattice.
Shanghai Jiao Tong University
3
0
7
Current video MLLMs struggle to grasp fleeting visual events, with top models barely surpassing 39% accuracy on critical momentary tasks.
SkillOpt transforms agent skill development into a reproducible optimization process, achieving state-of-the-art results by treating skills as trainable parameters.
Today's best AI agents can only solve 55% of real-world academic tasks that university students find challenging, revealing a significant gap between current AI capabilities and the demands of academic workflows.
