LongAttn: Selecting Long-context Training Data via Token-level Attention.
Longyun Wu, Dawei Zhu, Guangxiang Zhao, Zhuocheng Yu, Junfeng Ran, Xiangyu Wong, Lin Sun, Sujian Li
Browse the full ACL paper archive.
Longyun Wu, Dawei Zhu, Guangxiang Zhao, Zhuocheng Yu, Junfeng Ran, Xiangyu Wong, Lin Sun, Sujian Li
Browse the full ACL paper archive.