Repository of the paper "Accelerating Transformer Inference for Translation via Parallel Decoding"