mlki/ts@mlki bert-base-multilingual-cased
1 version Architecture: pfeiffer non-linearity: relu reduction factor: 16 Head: 

Knowledge adapter set for multilingual knowledge graph integration. This adapter is for factual triple enhancement (sentence-level). We trained it with triples from T-REx across 84 languages.

mlki/ts@mlki xlm-roberta-base
1 version Architecture: pfeiffer non-linearity: relu reduction factor: 16 Head: 

Knowledge adapter set for multilingual knowledge graph integration. This adapter is for factual triple enhancement (sentence-level). We trained it with triples from T-REx across 84 languages.

mlki/ts@mlki xlm-roberta-large
1 version Architecture: pfeiffer non-linearity: relu reduction factor: 16 Head: 

Knowledge adapter set for multilingual knowledge graph integration. This adapter is for factual triple enhancement (sentence-level). We trained it with triples from T-REx across 84 languages.