QUACKIE: A NLP Classification Task With Ground Truth Explanations
NLP Interpretability aims to increase trust in model predictions. This makes evaluating interpretability approaches a pressing issue. There are multiple datasets for evaluating NLP Interpretability, but their dependence on human provided ground truths raises questions about their unbiasedness. In this work, we take a different approach and formulate a specific classification task by diverting question-answering datasets. For this custom classification task, the interpretability ground-truth arises directly from the definition of the classification problem. We use this method to propose a benchmark and lay the groundwork for future research in NLP interpretability by evaluating a wide range of current state of the art methods.
Code (0)
등록된 구현이 없습니다.
Tasks
ClassificationGeneral ClassificationQuestion AnsweringMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Can We Trust Explainable AI Methods on ASR? An Evaluation on Phoneme Recognition
Explainable AI (XAI) techniques have been widely used to help explain and understand the output of deep learning models in fields such as image classification and Natural Language Processing. Interest in using XAI techni…
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Deep LearningExplainable Artificial Intelligence (XAI)+5Evaluating the Correctness of Explainable AI Algorithms for Classification
Explainable AI has attracted much research attention in recent years with feature attribution algorithms, which compute "feature importance" in predictions, becoming increasingly popular. However, there is little analysi…
Binary ClassificationClassificationExplainable Artificial Intelligence (XAI)Feature ImportanceEvaluating Model Explanations without Ground Truth
There can be many competing and contradictory explanations for a single model prediction, making it difficult to select which one to use. Current explanation evaluation frameworks measure quality by comparing against ide…
Feature ImportancemodelSensitivityEvaluating Explainability for Graph Neural Networks
As post hoc explanations are increasingly used to understand the behavior of graph neural networks (GNNs), it becomes crucial to evaluate the quality and reliability of GNN explanations. However, assessing the quality of…
Provable concept learning for interpretable predictions using variational autoencoders
In safety-critical applications, practitioners are reluctant to trust neural networks when no interpretable explanations are available. Many attempts to provide such explanations revolve around pixel-based attributions o…
Variational Inference