Jian Li, Xing Wang 0007, Zhaopeng Tu, Michael R. Lyu. On the diversity of multi-head attention. Neurocomputing, 454:14-24, 2021. [doi]
@article{LiWTL21, title = {On the diversity of multi-head attention}, author = {Jian Li and Xing Wang 0007 and Zhaopeng Tu and Michael R. Lyu}, year = {2021}, doi = {10.1016/j.neucom.2021.04.038}, url = {https://doi.org/10.1016/j.neucom.2021.04.038}, researchr = {https://researchr.org/publication/LiWTL21}, cites = {0}, citedby = {0}, journal = {Neurocomputing}, volume = {454}, pages = {14-24}, }