Instance-Based Learning of Span Representations: A Case Study through Named Entity Recognition
Preprint
Abstract
Interpretable rationales for model predictions play a critical role in practical applications. In this study, we develop models possessing interpretable inference process for structured prediction. Specifically, we present a method of instance-based learning that learns similarities between spans. At inference time, each span is assigned a class label based on its similar spans in the training set, where it is easy to understand how much each training instance contributes to the predictions. Through empirical analysis on named entity recognition, we demonstrate that our method enables to build models that have high interpretability without sacrificing performance.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- A Boundary-aware Neural Model for Nested Named Entity Recognition2019
- Deep Exhaustive Model for Nested Named Entity Recognition2018
- Neural Segmental Hypergraphs for Overlapping Mention Recognition2018
- Nested Named Entity Recognition Revisited2018
- Labeling Gaps Between Words: Recognizing Overlapping Mentions with Mention Separators2017
- End-to-end Neural Coreference Resolution2017
- Named Entity Recognition with Bidirectional LSTM-CNNsTransactions of the Association for Computational Linguistics, 2016
- Rationalizing Neural Predictions2016
- Joint Mention Extraction and Classification with Mention Hypergraphs2015
- GENIA corpus—a semantically annotated corpus for bio-textminingBioinformatics, 2003