关于解释和减轻NLU模型的学习行为
Towards Interpreting and Mitigating Shortcut Learning Behavior of NLU Models
摘要:
最近的研究表明,NLU模型倾向于依赖快捷功能进行预测,而无法实现真正的语言理解。因此,这些模型无法推广到真实世界的分布外数据。在这项工作中,我们证明了NLU训练集中的单词可以建模为长尾分布。有两个发现:1)NLU模型对位于长尾分布头部的特征有很强的偏好,2)在模型训练的早期几次迭代中,会提取出非线性特征。这两个观察进一步制定一个测量,可以量化每个训练样本的捷径程度。基于这种捷径度量,我们提出了一个捷径缓解框架LTGR,以抑制模型对具有大捷径度的样本进行过度自信的预测。三个NLU基准测试的实验结果表明,我们的长尾分布解释准确地反映了NLU模型的捷径学习行为。实验分析进一步表明,LTGR可以提高OOD数据的泛化精度,同时保持分布数据的精度。
英文摘要:
Recent studies indicate that NLU models are prone to rely on shortcut features for prediction, without achieving true language understanding. As a result, these models fail to generalize to real-world out-of-distribution data. In this work, we show that the words in the NLU training set can be modeled as a long-tailed distribution. There are two findings: 1) NLU models have strong preference for features located at the head of the long-tailed distribution, and 2) Shortcut features are picked up during very early few iterations of the model training. These two observations are further employed to formulate a measurement which can quantify the shortcut degree of each training sample. Based on this shortcut measurement, we propose a shortcut mitigation framework LTGR, to suppress the model from making overconfident predictions for samples with large shortcut degree. Experimental results on three NLU benchmarks demonstrate that our long-tailed distribution explanation accurately reflects the shortcut learning behavior of NLU models. Experimental analysis further indicates that LTGR can improve the generalization accuracy on OOD data, while preserving the accuracy on in-distribution data.
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。