# Stack-propagation: Improved Representation Learning for Syntax

arXiv: https://arxiv.org/abs/1603.06598 · HTML: https://mlsanity.com/paper/1603.06598/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Mar 2016.

## Built on by (most cited first, 7 of 7)

- [Structural Guidance for Transformer Language Models](https://mlsanity.com/paper/2108.00104/) (2021)
- [Better Neural Machine Translation by Extracting Linguistic Information from BERT](https://mlsanity.com/paper/2104.02831/) (2021)
- [Extending Word-Level Quality Estimation for Post-Editing Assistance](https://mlsanity.com/paper/2209.11378/) (2022)
- [Understanding Emotion Valence is a Joint Deep Learning Task](https://mlsanity.com/paper/2305.17422/) (2023)
- [Auxiliary Tasks to Boost Biaffine Semantic Dependency Parsing](https://mlsanity.com/paper/2402.07682/) (2024)
- [Quantity vs. Quality of Monolingual Source Data in Automatic Text Translation: Can It Be Too Little If It Is Too Good?](https://mlsanity.com/paper/2410.13783/) (2024)
- [A Stack-Propagation Framework for Low-Resource Personalized Dialogue Generation](https://mlsanity.com/paper/2410.20174/) (2024)

## Abstract (arXiv)

Traditional syntax models typically leverage part-of-speech (POS) information by constructing features from hand-tuned templates. We demonstrate that a better approach is to utilize POS tags as a regularizer of learned representations. We propose a simple method for learning a stacked pipeline of models which we call "stack-propagation". We apply this to dependency parsing and tagging, where we use the hidden layer of the tagger network as a representation of the input tokens for the parser. At test time, our parser does not require predicted POS tags. On 19 languages from the Universal Dependencies, our method is 1.3% (absolute) more accurate than a state-of-the-art graph-based approach and 2.7% more accurate than the most comparable greedy model.