Solving SCAN Tasks with Data Augmentation and Input Embeddings

Code

• github.com/michal-au/scan-around-twice
  OfficialIn papernone★ 0

Abstract

We address the compositionality challenge presented by the SCAN benchmark. Using data augmentation and a modification of the standard seq2seq architecture with attention, we achieve SOTA results on all the relevant tasks from the benchmark, showing the models can generalize to words used in unseen contexts. We propose an extension of the benchmark by a harder task, which cannot be solved by the proposed method.

Tasks

Reproductions