# An Imitation Learning Approach to Unsupervised Parsing

arXiv: https://arxiv.org/abs/1906.02276 · HTML: https://mlsanity.com/paper/1906.02276/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jun 2019.

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

- [Modeling Hierarchical Structures with Continuous Recursive Neural Networks](https://mlsanity.com/paper/2106.06038/) (2021)
- [Weakly Supervised Explainable Phrasal Reasoning with Neural Fuzzy Logic](https://mlsanity.com/paper/2109.08927/) (2021)
- [Ensemble Distillation for Unsupervised Constituency Parsing](https://mlsanity.com/paper/2310.01717/) (2023)
- [Weakly Supervised Reasoning by Neuro-Symbolic Approaches](https://mlsanity.com/paper/2309.13072/) (2023)
- [Increasing The Performance of Cognitively Inspired Data-Efficient Language Models via Implicit Structure Building](https://mlsanity.com/paper/2310.20589/) (2023)
- [The Emergence of Chunking Structures with Hierarchical RNN](https://mlsanity.com/paper/2309.04919/) (2023)
- [Tree-Averaging Algorithms for Ensemble-Based Unsupervised Discontinuous Constituency Parsing](https://mlsanity.com/paper/2403.00143/) (2024)

## Abstract (arXiv)

Recently, there has been an increasing interest in unsupervised parsers that optimize semantically oriented objectives, typically using reinforcement learning. Unfortunately, the learned trees often do not match actual syntax trees well. Shen et al. (2018) propose a structured attention mechanism for language modeling (PRPN), which induces better syntactic structures but relies on ad hoc heuristics. Also, their model lacks interpretability as it is not grounded in parsing actions. In our work, we propose an imitation learning approach to unsupervised parsing, where we transfer the syntactic knowledge induced by the PRPN to a Tree-LSTM model with discrete parsing actions. Its policy is then refined by Gumbel-Softmax training towards a semantically oriented objective. We evaluate our approach on the All Natural Language Inference dataset and show that it achieves a new state of the art in terms of parsing F-score, outperforming our base models, including the PRPN.