Seq2Edits: Sequence Transduction Using Span-level Edit Operations
This addresses efficiency and explainability issues in NLP tasks with overlapping input-output texts, though it is incremental as it builds on existing sequence-to-sequence methods.
The paper tackles sequence transduction tasks in NLP by proposing Seq2Edits, which uses span-level edit operations to represent changes, resulting in competitive performance across five tasks and up to 5.2x faster inference for grammatical error correction.
We propose Seq2Edits, an open-vocabulary approach to sequence editing for natural language processing (NLP) tasks with a high degree of overlap between input and output texts. In this approach, each sequence-to-sequence transduction is represented as a sequence of edit operations, where each operation either replaces an entire source span with target tokens or keeps it unchanged. We evaluate our method on five NLP tasks (text normalization, sentence fusion, sentence splitting & rephrasing, text simplification, and grammatical error correction) and report competitive results across the board. For grammatical error correction, our method speeds up inference by up to 5.2x compared to full sequence models because inference time depends on the number of edits rather than the number of target tokens. For text normalization, sentence fusion, and grammatical error correction, our approach improves explainability by associating each edit operation with a human-readable tag.