tasksource/ModernBERT-nli demonstration

This space uses tasksource/ModernBERT-base-nli, fine-tuned from answerdotai/ModernBERT-base on tasksource classification tasks. This NLI model achieves high accuracy on categorization, logical reasoning and long-context NLI, outperforming Llama 3 8B on ConTRoL (long-context NLI) and FOLIO (logical reasoning).