NLIZE: A Perturbation-Driven Visual Interrogation Tool for Analyzing and Interpreting Natural Language Inference Models
NLIZE: A Perturbation-Driven Visual Interrogation Tool for Analyzing and Interpreting Natural Language Inference Models
复制标题
DOI：
10.1109/tvcg.2018.2865230
复制
发表时间：
2019-01
影响因子：
5.2
通讯作者：
Shusen Liu;Zhimin Li;Tao Li;Vivek Srikumar;Valerio Pascucci;P. Bremer
中科院分区：
文献类型：
--
作者：
Shusen Liu;Zhimin Li;Tao Li;Vivek Srikumar;Valerio Pascucci;P. Bremer
随着深度学习的最新进展，神经网络模型在自然语言处理中的许多语言任务中获得了最先进的性能。然而，这种快速的进步也带来了巨大的挑战。神经网络模型的不透明性导致了难以调试的系统和难以解释的机制。在这里，我们介绍了一个可视化系统，通过可视化元素和底层模型之间紧密而灵活的集成，允许用户通过扰动输入，内部状态和预测来询问模型，同时观察管道其他部分的变化。我们使用自然语言推理问题作为一个例子来说明扰动驱动的范式可以帮助领域专家评估模型的潜在限制，探测其内部状态，并解释和形成有关基本模型机制（如注意力）的假设。
With the recent advances in deep learning, neural network models have obtained state-of-the-art performances for many linguistic tasks in natural language processing. However, this rapid progress also brings enormous challenges. The opaque nature of a neural network model leads to hard-to-debug-systems and difficult-to-interpret mechanisms. Here, we introduce a visualization system that, through a tight yet flexible integration between visualization elements and the underlying model, allows a user to interrogate the model by perturbing the input, internal state, and prediction while observing changes in other parts of the pipeline. We use the natural language inference problem as an example to illustrate how a perturbation-driven paradigm can help domain experts assess the potential limitation of a model, probe its inner states, and interpret and form hypotheses about fundamental model mechanisms such as attention.