SIGMORPHON–UniMorph 2022 Shared Task 0: Modeling Inflection in Language Acquisition

Kodner, Jordan; Khalifa, Salam

doi:10.18653/v1/2022.sigmorphon-1.18

Cited by 7 publications

(9 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…OSU entered systems for both Part 1 (multilingual inflection; Kodner et al (2022)) and Part 2 (learning trajectories: Kodner and Khalifa (2022)). However, we did not attempt all parts of the Part 1 task.…”

Section: Resultsmentioning

confidence: 99%

OSU at SigMorphon 2022: Analogical Inflection With Rule Features

Elsner¹,

Court²

2022

Proceedings of the 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

View full text Add to dashboard Cite

OSU's inflection system is a transformer whose input is augmented with an analogical exemplar showing how to inflect a different word into the target cell. In addition, alignment-based heuristic features indicate how well the exemplar is likely to match the output. OSU's scores substantially improve over the baseline transformer for instances where an exemplar is available, though not quite matching the challenge winner. In Part 2, the system shows a tendency to over-apply the majority pattern in English, but not Arabic.

show abstract

Section: Resultsmentioning

confidence: 99%

OSU at SigMorphon 2022: Analogical Inflection With Rule Features

Elsner¹,

Court²

2022

Proceedings of the 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

View full text Add to dashboard Cite

show abstract

“…The details of the task description can be found at https://github.com/sigmorphon/ 2022InflectionST. We use the data provided by the SIGMORPHON 2022 shared task (Part 2) (Kodner and Khalifa, 2022). The data features lemmas, inflections, and corresponding morphosyntactic description (MSD) using the uni-morph schema (Kirov et al, 2018).…”

Section: Background and Datamentioning

confidence: 99%

HeiMorph at SIGMORPHON 2022 Shared Task on Morphological Acquisition Trajectories

Ramarao¹,

Zinova²,

Tang³

et al. 2022

Proceedings of the 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

View full text Add to dashboard Cite

This paper presents the submission by the HeiMorph team to the SIGMORPHON 2022 task 2 of Morphological Acquisition Trajectories. Across all experimental conditions, we have found no evidence for the so-called Ushaped development trajectory. Our submitted systems achieve an average test accuracies of 55.5% on Arabic, 67% on German and 73.38% on English. We found that, bigram hallucination provides better inferences only for English and Arabic and only when the number of hallucinations remains low.

show abstract

“…Input Output SEGM hierarchisms hierarch @@y @@ism @@s INFL sue V;PST sued 1. Typologically diverse morphological inflection (Kodner et al, 2022) 2. Morphological acquisition trajectories (Kodner and Khalifa, 2022) All our submissions rely on the same neural hardattention transducer architecture that has shown strong language-independent performance in a variety of character-level transduction tasks in morphology, grapheme-to-phoneme conversion, and text normalization Clematide, 2018, 2020a,b).…”

Section: Taskmentioning

confidence: 99%

CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation

Wehrli¹,

Clematide²,

Makarov³

2022

Proceedings of the 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

View full text Add to dashboard Cite

This paper describes the submissions of the team of the Department of Computational Linguistics, University of Zurich, to the SIGMOR-PHON 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 well-suited 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.

show abstract

SIGMORPHON–UniMorph 2022 Shared Task 0: Modeling Inflection in Language Acquisition

Cited by 7 publications

References 34 publications

OSU at SigMorphon 2022: Analogical Inflection With Rule Features

OSU at SigMorphon 2022: Analogical Inflection With Rule Features

HeiMorph at SIGMORPHON 2022 Shared Task on Morphological Acquisition Trajectories

CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation

Contact Info

Product

Resources

About