Poolingformer: Long Document Modeling with Pooling Attention. (arXiv:2105.04371v2 [cs.CL] UPDATED)

In this paper, we introduce a two-level attention schema, Poolingformer, for
long document modeling. Its first level uses a smaller sliding window pattern
to aggregate information from neighbors. Its second level employs a larger
window to increase receptive fields with pooling attention to reduce both
computational cost and memory consumption. We first evaluate Poolingformer on
two long sequence QA tasks: the monolingual NQ and the multilingual TyDi QA.
Experimental results show that Poolingformer sits atop three official
leaderboards measured by …

arxiv attention modeling pooling