HeRo: RoBERTa and Longformer Hebrew Language Models
This provides Hebrew NLP practitioners with improved tools, but it is incremental as it adapts existing methods to a new language.
The authors tackled the lack of Hebrew NLP resources by creating HeRo and LongHeRo models, achieving state-of-the-art performance on tasks like sentiment analysis and document classification.
In this paper, we fill in an existing gap in resources available to the Hebrew NLP community by providing it with the largest so far pre-train dataset HeDC4, a state-of-the-art pre-trained language model HeRo for standard length inputs and an efficient transformer LongHeRo for long input sequences. The HeRo model was evaluated on the sentiment analysis, the named entity recognition, and the question answering tasks while the LongHeRo model was evaluated on the document classification task with a dataset composed of long documents. Both HeRo and LongHeRo presented state-of-the-art performance. The dataset and model checkpoints used in this work are publicly available.