Inference, especially those derived from inductive processes, is a crucial component in our conversation to complement the information implicitly or explicitly conveyed by a speaker. While recent large language models show remarkable advances in inference tasks, their performance in inductive reasoning, where not all information is present in the context, is far behind deductive reasoning. In this paper, we analyze the behavior of the models based on the task difficulty defined by the semantic information gap -- which distinguishes inductive and deductive reasoning (Johnson-Laird, 1988, 1993). Our analysis reveals that the disparity in information between dialogue contexts and desired inferences poses a significant challenge to the inductive inference process. To mitigate this information gap, we investigate a contrastive learning approach by feeding negative samples. Our experiments suggest negative samples help models understand what is wrong and improve their inference generations.

翻译：推理，尤其是源自归纳过程的部分，是对话中补充说话者隐式或显式传达信息的关键组成部分。尽管近期的大语言模型在推理任务上展现出显著进步，但其在归纳推理（即上下文未提供全部信息的情况）中的表现远不及演绎推理。本文基于语义信息鸿沟（该概念区分了归纳推理与演绎推理，参见Johnson-Laird，1988, 1993）所定义的任务难度，分析了模型的行为。我们的分析表明，对话语境与期望推理之间的信息差异对归纳推理过程构成了重大挑战。为弥合这一信息鸿沟，我们研究了一种通过输入负样本的对比学习方法。实验结果表明，负样本有助于模型理解错误所在，并改进其推理生成质量。