Recognizing Textual Entailment Benchmark | The Regularizer