InfoTabS

该数据集旨在评估语言模型在面对不同输入扰动时的表现，特别关注于表格型自然语言推理任务（Tabular-NLI）。它被用于展示模型性能如何受到各种扰动的影响，包括否定挑战和不同的训练策略。该数据集的任务是自然语言推理（NLI）。

This dataset is designed to evaluate the performance of language models under diverse input perturbations, with a particular focus on the Tabular Natural Language Inference (Tabular-NLI) task. It is employed to illustrate how various perturbations—including negation challenges and different training strategies—impact model performance. The core task of this dataset is Natural Language Inference (NLI).