Papers by Michael Ginn
Resisting the Lure of the Skyline: Grounding Practices in Active Learning for Morphological Inflection (2024.acl-short)
Copied to clipboard
| Challenge: | Several approaches to active learning are available, including confidence-based, diversity-based and committee-based. |
| Approach: | They propose to use a baseline and a skyline to measure the accuracy of the unannotated sample pool. |
| Outcome: | The proposed model outperforms a random selection baseline and a skyline approach. |
Neural Induction of Finite-State Transducers (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to construct finite-state transducers by hand are difficult and require domain knowledge and significant human effort. |
| Approach: | They propose a method for automatically constructing unweighted FSTs following the hidden state geometry learned by a recurrent neural network. |
| Outcome: | The proposed method outperforms classical transducer learning algorithms by up to 87% accuracy on held-out test sets. |
Massively Multilingual Joint Segmentation and Glossing (2026.acl-long)
Copied to clipboard
Michael Ginn, Lindia Tjuatja, Enora Rice, Ali Marashian, Maria Valentini, Jasmine Xu, Graham Neubig, Alexis Palmer
| Challenge: | Existing models generate morpheme-level glosses but assign them to whole words without predicting the actual morphological boundaries, making them less interpretable and therefore untrustworthy to human annotators. |
| Approach: | They propose to use neural networks to predict interlinear glosses and morphological segmentation from raw text. |
| Outcome: | The proposed model outperforms GlossLM on glossing and beats open-source models on segmentation, glossing, and alignment. |
Can we teach language models to gloss endangered languages? (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Prior research has explored statistical and neural methods for automatically producing IGT. |
| Approach: | They propose to use in-context learning to generate interlinear glossed text . they propose to employ supervised learning to select examples to provide in-text . |
| Outcome: | The proposed methods beat standard transformer baselines, despite requiring no training at all. |
GlossLM: A Massively Multilingual Corpus and Pretrained Model for Interlinear Glossed Text (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing resources for standardized, easily accessible IGT data limit their applicability to linguistic research. |
| Approach: | They compile the largest existing corpus of interlinear glossed text data from a variety of sources and use it to generate annotated text. |
| Outcome: | The proposed model outperforms SOTA models on monolingual corpora by 6.6%. |
PyFoma: a Python finite-state compiler module (2024.acl-demos)
Copied to clipboard
| Challenge: | Finite-state models can be used to constrain output of neural networks to prevent text generation that fails to adhere to a specific format. |
| Approach: | They propose to build finite-state automata from regular expressions, string rewriting rules, right-linear grammars, or low-level state/transition manipulation. |
| Outcome: | The module is designed for teaching finite-state models and finite models. |
Is linguistically-motivated data augmentation worth it? (2025.acl-long)
Copied to clipboard
| Challenge: | Data augmentation is a widely-used technique for addressing data scarcity. |
| Approach: | They compare linguistically-motivated and linguisticly-naive data augmentation strategies for two low-resource languages with different morphological properties. |
| Outcome: | The proposed methods produce synthetic data that follows all linguistic constraints, but they require linguistic expertise and are more difficult to implement. |