NeoDictaBERT: Pushing the Frontier of BERT models for Hebrew
This provides improved models for Hebrew NLP research and applications, though it is incremental as it adapts an existing architecture to a specific language.
The authors tackled the problem of outdated BERT architectures for Hebrew NLP by introducing NeoDictaBERT and NeoDictaBERT-bilingual, which outperform existing models on almost all Hebrew benchmarks and show strong results on retrieval tasks.
Since their initial release, BERT models have demonstrated exceptional performance on a variety of tasks, despite their relatively small size (BERT-base has ~100M parameters). Nevertheless, the architectural choices used in these models are outdated compared to newer transformer-based models such as Llama3 and Qwen3. In recent months, several architectures have been proposed to close this gap. ModernBERT and NeoBERT both show strong improvements on English benchmarks and significantly extend the supported context window. Following their successes, we introduce NeoDictaBERT and NeoDictaBERT-bilingual: BERT-style models trained using the same architecture as NeoBERT, with a dedicated focus on Hebrew texts. These models outperform existing ones on almost all Hebrew benchmarks and provide a strong foundation for downstream tasks. Notably, the NeoDictaBERT-bilingual model shows strong results on retrieval tasks, outperforming other multilingual models of similar size. In this paper, we describe the training process and report results across various benchmarks. We release the models to the community as part of our goal to advance research and development in Hebrew NLP.