HalleluBERT released for advanced Hebrew NLP tasks

Researchers have developed HalleluBERT, a new family of RoBERTa-based encoders specifically for the Hebrew language. Trained on a substantial corpus of Hebrew text, HalleluBERT has demonstrated superior performance on native Hebrew benchmarks for named entity recognition and sentiment classification compared to existing models. The researchers are releasing the model weights and tokenizer under an MIT license to foster reproducible research in Hebrew NLP. AI

IMPACT Enables more advanced NLP applications and research specifically for the Hebrew language.

RANK_REASON The cluster contains an academic paper detailing a new model release for a specific language. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• HalleluBERT
• Hebrew
• Raphael Schmitt
• RoBERTa

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →