Dynamic Top-k Estimation Consolidates Disagreement between Feature Attribution Methods
Feature attribution scores are used for explaining the prediction of a text classifier to users by highlighting a k number of tokens. In this work, we propose a way to determine the number of optimal k tokens that should be displayed from sequential properties of the attribution scores. Our approach is dynamic across sentences, method-agnostic, and deals with sentence length bias. We compare agreement between multiple methods and humans on an NLI task, using fixed k and dynamic k. We find that perturbation-based methods and Vanilla Gradient exhibit highest agreement on most method--method and method--human agreement metrics with a static k. Their advantage over other methods disappears with dynamic ks which mainly improve Integrated Gradient and GradientXInput. To our knowledge, this is the first evidence that sequential properties of attribution scores are informative for consolidating attribution signals for human interpretation.
Code (0)
등록된 구현이 없습니다.
Tasks
SentenceSimilar Papers 제목 키워드 기반
ConR: Contrastive Regularizer for Deep Imbalanced Regression
Imbalanced distributions are ubiquitous in real-world data. They create constraints on Deep Neural Networks to represent the minority labels and avoid bias towards majority labels. The extensive body of imbalanced approa…
Deep imbalanced regressionregressionDual-Stream Collaborative Transformer for Image Captioning
Current region feature-based image captioning methods have progressed rapidly and achieved remarkable performance. However, they are still prone to generating irrelevant descriptions due to the lack of contextual informa…
Image CaptioningUfM*: Uncertainty from Motion* for DNN Depth Estimation Using Gaussians
Reliable uncertainty estimation is critical for deploying monocular depth deep neural networks (DNNs) in safety-critical robotic systems. Conventional uncertainty methods such as ensembles and sampling-based approaches r…
Depth EstimationLearning semantical dynamics and spatiotemporal collaboration for human pose estimation in video
Temporal modeling and spatio-temporal collaboration are pivotal techniques for video-based human pose estimation. Most state-of-the-art methods adopt optical flow or temporal difference, learning local visual content cor…
Optical Flow EstimationPose EstimationAETTA: Label-Free Accuracy Estimation for Test-Time Adaptation
Test-time adaptation (TTA) has emerged as a viable solution to adapt pre-trained models to domain shifts using unlabeled test data. However, TTA faces challenges of adaptation failures due to its reliance on blind adapta…
PredictionTest-time Adaptation