1 paper · 1 filter
Ziyuan He, Yuxuan Wang, Jiaqi Li +2
Large language models (LLMs) are equipped with increasingly extended context windows recently, yet their long context understanding capabilities over long dependency tasks remain f…