Joint Parsing and Generation for Abstractive Summarization
This addresses the issue of grammaticality and faithfulness in abstractive summarization for NLP applications, though it is incremental as it builds on existing neural methods.
The paper tackles the problem of ungrammatical and meaning-distorting sentences in abstractive summarization by jointly generating a summary sentence and its syntactic dependency parse, demonstrating competitive results on multiple datasets.
Sentences produced by abstractive summarization systems can be ungrammatical and fail to preserve the original meanings, despite being locally fluent. In this paper we propose to remedy this problem by jointly generating a sentence and its syntactic dependency parse while performing abstraction. If generating a word can introduce an erroneous relation to the summary, the behavior must be discouraged. The proposed method thus holds promise for producing grammatical sentences and encouraging the summary to stay true-to-original. Our contributions of this work are twofold. First, we present a novel neural architecture for abstractive summarization that combines a sequential decoder with a tree-based decoder in a synchronized manner to generate a summary sentence and its syntactic parse. Secondly, we describe a novel human evaluation protocol to assess if, and to what extent, a summary remains true to its original meanings. We evaluate our method on a number of summarization datasets and demonstrate competitive results against strong baselines.