Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

An example of neural NER with sequence labeling.

More »

Fig 1 Expand

Fig 2.

The internal structure of a one-layer transformer.

(a) Standard Transformer. (b) Transformer with Adapters.

More »

Fig 2 Expand

Fig 3.

The pretrained transformer-based language model coupling with adapters and PGN.

More »

Fig 3 Expand

Table 1.

Data statistics, where the number of sentences and entities are reported, and Devel indicates the development set.

More »

Table 1 Expand

Table 2.

Main results of single-source cross-lingual NER, where lavg indicates the averaged performance for each target language, and avg denotes the overall average F-scores of all source-target pairs.

More »

Table 2 Expand

Table 3.

Main results of multi-source cross-lingual NER, where all other languages except the target language itself are exploited as the source languages.

More »

Table 3 Expand

Table 4.

Comparisons with previous studies.

More »

Table 4 Expand

Fig 4.

An example of word alignment visualization between a German sentence and its English translation, where the solid arrows are gold-standard being all correctly predicted by XLM, and the dashed arrows are incorrectly aligned by mBERT, and the others are the same for the two types of word representations.

More »

Fig 4 Expand

Table 5.

The comparisons between the fine-tuning and feature-based adapter exploration methods, where XLM is used as the input language model.

More »

Table 5 Expand

Fig 5.

The similarity heatmap of language embeddings for different language pairs, where deeper color indicates higher similarity.

More »

Fig 5 Expand

Table 6.

A case study, where the text with underlines indicates errors.

More »

Table 6 Expand