# Cross-Register Projection for Headline Part of Speech Tagging

arXiv: https://arxiv.org/abs/2109.07483 · HTML: https://mlsanity.com/paper/2109.07483/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 54 references that are in the map.
- First posted on arXiv 15 Sep 2021.

## Builds on (most cited first, 13 of 13)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Neural Architectures for Named Entity Recognition](https://mlsanity.com/paper/1603.01360/) (2016)
- [A Neural Attention Model for Abstractive Sentence Summarization](https://mlsanity.com/paper/1509.00685/) (2015)
- [Positional Encoding to Control Output Sequence Length](https://mlsanity.com/paper/1904.07418/) (2019)
- [Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations](https://mlsanity.com/paper/1603.04351/) (2016)
- [Universal Dependency Parsing from Scratch](https://mlsanity.com/paper/1901.10457/) (2019)
- [GoodNewsEveryone: A Corpus of News Headlines Annotated with Emotions, Semantic Roles, and Reader Perception](https://mlsanity.com/paper/1912.03184/) (2019)
- [Parsing Tweets into Universal Dependencies](https://mlsanity.com/paper/1804.08228/) (2018)
- [News Headline Grouping as a Challenging NLU Task](https://mlsanity.com/paper/2105.05391/) (2021)
- [Cross-Lingual Morphological Tagging for Low-Resource Languages](https://mlsanity.com/paper/1606.04279/) (2016)
- [Multi-task Domain Adaptation for Sequence Tagging](https://mlsanity.com/paper/1608.02689/) (2016)
- [Analysing Errors of Open Information Extraction Systems](https://mlsanity.com/paper/1707.07499/) (2017)
- [NSTM: Real-Time Query-Driven News Overview Composition at Bloomberg](https://mlsanity.com/paper/2006.01117/) (2020)

## Built on by (most cited first, 1 of 1)

- [Weakly Supervised Headline Dependency Parsing](https://mlsanity.com/paper/2301.10371/) (2023)

## Abstract (arXiv)

Part of speech (POS) tagging is a familiar NLP task. State of the art taggers routinely achieve token-level accuracies of over 97% on news body text, evidence that the problem is well understood. However, the register of English news headlines, "headlinese", is very different from the register of long-form text, causing POS tagging models to underperform on headlines. In this work, we automatically annotate news headlines with POS tags by projecting predicted tags from corresponding sentences in news bodies. We train a multi-domain POS tagger on both long-form and headline text and show that joint training on both registers improves over training on just one or naively concatenating training sets. We evaluate on a newly-annotated corpus of over 5,248 English news headlines from the Google sentence compression corpus, and show that our model yields a 23% relative error reduction per token and 19% per headline. In addition, we demonstrate that better headline POS tags can improve the performance of a syntax-based open information extraction system. We make POSH, the POS-tagged Headline corpus, available to encourage research in improved NLP models for news headlines.