HyperAI

Morpheme Segmentaiton On Unimorph 4 0

Metrics

f1 macro avg (subtask 2)
lev dist (subtask 2)
macro avg (subtask 1)

Results

Performance results of various models on this benchmark

Model Name
f1 macro avg (subtask 2)
lev dist (subtask 2)
macro avg (subtask 1)
Paper TitleRepository
WordPiece40.5917.5415.89The SIGMORPHON 2022 Shared Task on Morpheme Segmentation
BiLSTM for seq labelling (Tü_Seg-1)--96.06SIGMORPHON 2022 Shared Task on Morpheme Segmentation Submission Description: Sequence Labelling for Word-Level Morpheme Segmentation-
CLUZH-187.685.69-CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation
Char LSTM (DeepSPIN-2; soft-attention, 1-5 entmax)--97.15Beyond Characters: Subword-level Morpheme Segmentation-
HMM (JB132; trained w EM)--58.39JB132 submission to the SIGMORPHON 2022 Shared Task 3 on Morphological Segmentation-
Ensemble of hard-attention transducers (CLUZH)--96.85CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation
AUUH_C70.7635.94-Morfessor-enriched features and multilingual training for canonical morphological segmentation-
Bidirectional GRU + Morfessor features (AUUH_F)66.7336.3593.72Morfessor-enriched features and multilingual training for canonical morphological segmentation-
BiLSTM for seq labelling (Tü_Seg-2)82.074.71-SIGMORPHON 2022 Shared Task on Morpheme Segmentation Submission Description: Sequence Labelling for Word-Level Morpheme Segmentation-
Morfessor250.6512.0825.57The SIGMORPHON 2022 Shared Task on Morpheme Segmentation
AUUH_E73.2131.05-Morfessor-enriched features and multilingual training for canonical morphological segmentation-
ULM45.9914.2820.61The SIGMORPHON 2022 Shared Task on Morpheme Segmentation
CLUZH-388.145.58-CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation
Subword-ULM transformer (DeepSPIN-3; soft-attention, 1-5 entmax)--97.29Beyond Characters: Subword-level Morpheme Segmentation-
AUUH_D72.7536.38-Morfessor-enriched features and multilingual training for canonical morphological segmentation-
Char LSTM (DeepSPIN-1; soft-attention)--96.32Beyond Characters: Subword-level Morpheme Segmentation-
CLUZH-287.935.62-CLUZH at SIGMORPHON 2022 Shared Tasks on Morpheme Segmentation and Inflection Generation
AUUH_B89.773.50-Morfessor-enriched features and multilingual training for canonical morphological segmentation-
AUUH_A89.004.08-Morfessor-enriched features and multilingual training for canonical morphological segmentation-
0 of 19 row(s) selected.