Papers › CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation
CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation
Silvan Wehrli, Simon Clematide, Peter Makarov
This paper describes the submissions of the team of the Department of Computational Linguistics, University of Zurich, to the SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation. Our submissions use a character-level neural transducer that operates over traditional edit actions. While this model has been found particularly wellsuited for low-resource settings, using it with large data quantities has been difficult. Existing implementations could not fully profit from GPU acceleration and did not efficiently implement mini-batch training, which could be tricky for a transition-based system. For this year’s submission, we have ported the neural transducer to PyTorch and implemented true mini-batch training. This has allowed us to successfully scale the approach to large data quantities and conduct extensive experimentation. We report competitive results for morpheme segmentation (including sharing first place in part 2 of the challenge). We also demonstrate that reducing sentence-level morpheme segmentation to a word-level problem is a simple yet effective strategy. Additionally, we report strong results in inflection generation (the overall best result for large training sets in part 1, the best results in low-resource learning trajectories in part 2). Our code is publicly available.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
1 archive task tag without a task page not shown.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Morpheme Segmentaiton | UniMorph 4.0 | Ensemble of hard-attention transducers (CLUZH) | macro avg (subtask 1) | 96.85 | #3 of 19 | Archive leaderboard | report |
| Morpheme Segmentaiton | UniMorph 4.0 | CLUZH-3 | f1 macro avg (subtask 2) | 88.14 | #13 of 19 | Archive leaderboard | report |
| Morpheme Segmentaiton | UniMorph 4.0 | CLUZH-3 | lev dist (subtask 2) | 5.58 | #13 of 19 | Archive leaderboard | report |
| Morpheme Segmentaiton | UniMorph 4.0 | CLUZH-2 | f1 macro avg (subtask 2) | 87.93 | #14 of 19 | Archive leaderboard | report |
| Morpheme Segmentaiton | UniMorph 4.0 | CLUZH-2 | lev dist (subtask 2) | 5.62 | #14 of 19 | Archive leaderboard | report |
| Morpheme Segmentaiton | UniMorph 4.0 | CLUZH-1 | f1 macro avg (subtask 2) | 87.68 | #15 of 19 | Archive leaderboard | report |
| Morpheme Segmentaiton | UniMorph 4.0 | CLUZH-1 | lev dist (subtask 2) | 5.69 | #15 of 19 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections