Forming Trees with Treeformers

Popular models such as Transformers and LSTMs use tokens as its unit of information. That is, each token is encoded into a vector representation, and those vectors are used directly in a computation. However, humans frequently consider spans of tokens (i.e., phrases) instead of their constituent tokens. In this paper we introduce Treeformer, an architecture inspired by the CKY algorithm and Transformer which learns a composition operator and pooling function in order to construct hierarchical encodings for phrases and sentences. Our extensive experiments demonstrate the benefits of incorporating a hierarchical structure into the Transformer, and show significant improvements compared to a baseline Transformer in machine translation, abstractive summarization, and various natural language understanding tasks.