Follow
Picking the Underused Heads: A Network Pruning Perspective of Attention Head Selection for Fusing Dialogue Coreference Information. (arXiv:2312.09541v1 [cs.CL])
Picking the Underused Heads: A Network Pruning Perspective of Attention Head Selection for Fusing Dialogue Coreference Information. (arXiv:2312.09541v1 [cs.CL])