从充分性理性学习：分析解释忠实性与词级正则化策略之间的关系
Learning from Sufficient Rationales: Analysing the Relationship Between Explanation Faithfulness and Token-level Regularisation Strategies
AI总结：
研究探讨了理性充分性与词级正则化策略的关系，发现高信息量理性对分类影响有限，而充分性反映非理性上下文对分类的影响，且跨领域分类效果因任务和模型类型而异。
摘要：
自然语言的解释，即理性，是一种工具，用于评估模型是否为标签学习了正确的理由，还是依赖于数据集特定的捷径。充分性是估计理性信息量的常见指标，但其对理性信息对模型性能的影响作用有限。我们通过将充分性与两种建模范式联系起来来解决这一限制：模型识别理性中哪些词是组成部分的能力（通过词分类）以及通过在输入中纳入理性来提高模型性能的能力（通过注意力正则化）。我们发现，信息量大的理性不太可能帮助正确分类实例。相反，充分性捕捉了非理性化上下文对分类的影响，这种上下文在同一输入中干扰了理性信息。我们还发现，在模型输入中纳入理性信息可以提升跨领域分类，但结果在不同任务和模型类型中不一致。最后，充分性和词分类似乎无关。这些结果展示了理性的复杂性，表明能够系统捕捉此类信息的指标值得进一步研究。
英文摘要：
Human explanations of natural language, rationales, form a tool to assess whether models learn a label for the right reasons or rely on dataset-specific shortcuts. Sufficiency is a common metric for estimating the informativeness of rationales, but it provides limited insight into the effects of rationale information on model performance. We address this limitation by relating sufficiency to two modelling paradigms: the ability of models to identify which tokens are part of the rationale (through token classification) and the ability of improving model performance by incorporating rationales in the input (through attention regularisation). We find that highly informative rationales are not likely to help classify the instance correctly. Sufficiency conversely captures the classification impact of the non-rationalised context, which interferes with rationale information in the same input. We also find that incorporating rationale information in model inputs can boost cross-domain classification, but results are inconsistent per task and model type. Finally, sufficiency and token classification appear to be unrelated. These results exemplify the complexity of rationales, showing that metrics capable of systematically capturing this type of information merit further investigation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
