Augmented Natural Language for Generative Sequence Labeling
This work addresses the problem of multi-task sequence labeling for NLP researchers and practitioners, offering a novel generative approach that is incremental over prior discriminative methods.
The authors tackled joint sequence labeling and sentence-level classification by proposing a generative framework that uses a shared natural language output space, achieving new state-of-the-art results in few-shot slot labeling (e.g., 90.9% vs. 75.0% in 5-shot) and large improvements in low-resource settings (e.g., 63.83% vs. 46.27% over a BERT baseline).
We propose a generative framework for joint sequence labeling and sentence-level classification. Our model performs multiple sequence labeling tasks at once using a single, shared natural language output space. Unlike prior discriminative methods, our model naturally incorporates label semantics and shares knowledge across tasks. Our framework is general purpose, performing well on few-shot, low-resource, and high-resource tasks. We demonstrate these advantages on popular named entity recognition, slot labeling, and intent classification benchmarks. We set a new state-of-the-art for few-shot slot labeling, improving substantially upon the previous 5-shot ($75.0\% \rightarrow 90.9\%$) and 1-shot ($70.4\% \rightarrow 81.0\%$) state-of-the-art results. Furthermore, our model generates large improvements ($46.27\% \rightarrow 63.83\%$) in low-resource slot labeling over a BERT baseline by incorporating label semantics. We also maintain competitive results on high-resource tasks, performing within two points of the state-of-the-art on all tasks and setting a new state-of-the-art on the SNIPS dataset.