PoSE: Efficient Context Window Extension of LLMs via Positional Skip-wise Training.
Dawei Zhu, Nan Yang, Liang Wang, Yifan Song, Wenhao Wu, Furu Wei, Sujian Li
Browse the full ICLR paper archive.
Dawei Zhu, Nan Yang, Liang Wang, Yifan Song, Wenhao Wu, Furu Wei, Sujian Li
Browse the full ICLR paper archive.