# Imitation Learning for Neural Morphological String Transduction

arXiv: https://arxiv.org/abs/1808.10701 · HTML: https://mlsanity.com/paper/1808.10701/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Aug 2018.

## Built on by (most cited first, 10 of 10)

- [The SIGMORPHON 2022 Shared Task on Morpheme Segmentation](https://mlsanity.com/paper/2206.07615/) (2022)
- [Can a Transformer Pass the Wug Test? Tuning Copying Bias in Neural Morphological Inflection Models](https://mlsanity.com/paper/2104.06483/) (2021)
- [Frequency matters: Modeling irregular morphological patterns in Spanish with Transformers](https://mlsanity.com/paper/2410.21013/) (2024)
- [Morphological Processing of Low-Resource Languages: Where We Are and What's Next](https://mlsanity.com/paper/2203.08909/) (2022)
- [An Investigation of Noise in Morphological Inflection](https://mlsanity.com/paper/2305.16581/) (2023)
- [Computational Morphology with Neural Network Approaches](https://mlsanity.com/paper/2105.09404/) (2021)
- [Mind Your Moras: Orthography-Aware Error Analysis of Neural Japanese Morphological Generation](https://mlsanity.com/paper/2605.20043/) (2026)
- [Exploiting All Samples in Low-Resource Sentence Classification: Early Stopping and Initialization Parameters](https://mlsanity.com/paper/2111.06971/) (2021)
- [Improving Low-Resource Morphological Inflection via Self-Supervised Objectives](https://mlsanity.com/paper/2506.05227/) (2025)
- [Not All Irregularity Is Equal: Causally Isolating a Rare Failure Mode in Japanese Morphological Inflection](https://mlsanity.com/paper/2609.21179/) (2026)

## Abstract (arXiv)

We employ imitation learning to train a neural transition-based string transducer for morphological tasks such as inflection generation and lemmatization. Previous approaches to training this type of model either rely on an external character aligner for the production of gold action sequences, which results in a suboptimal model due to the unwarranted dependence on a single gold action sequence despite spurious ambiguity, or require warm starting with an MLE model. Our approach only requires a simple expert policy, eliminating the need for a character aligner or warm start. It also addresses familiar MLE training biases and leads to strong and state-of-the-art performance on several benchmarks.