Longer Attention Span: Increasing Transformer Context Length with Sparse Graph Processing Techniques

Re-assign community