TOWARDS HIERARCHICAL IMPORTANCE ATTRIBU-

By Xisen Jin. 2019

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Model Architectures

Citations in the corpus

SELFEXPLAIN: A Self-Explaining Architecture for Neural Text Classifiers (2021, methodology): “Trustability: In addition to plausibility, we also evaluate user trust of the explanations (Singh et al., 2019; Jin et al., 2020).”
Refining Language Models with Compositional Explanations (2021, methodology): “We first use the source model fS to make predictions on a small set of unlabeled instances randomly sampled from DT and present these post-hoc explanation heatmaps [19] to human annotators (see Fig.”
How Do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking (2020, methodology): “Their hierarchical versions (e.g., Singh et al. (2018); Jin et al. (2020)) also make a strong assumption about the structure of interaction (e.g., forming a tree) which may affect their faithfulness.”
Handling Bias in Toxic Speech Detection: A Survey (2022, methodology): “Using Sampling and Occlusion [47] as an explanation algorithm, the authors also validated a reduction of the importance score given by the regularized model to the identity terms, suggesting a decrease in lexical bias.”
Contextualizing Hate Speech Classifiers with Post-hoc Explanation (2020, methodology): “Our approach to this problem is to use the Sampling and Occlusion (SOC) explanation algorithm, which estimates model-agnostic, post-hoc feature importance (Jin et al., 2020).”
Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking (2020, methodology): “Outside of graph-speciﬁc methods, one line of research involves decomposing the output into a part attributed to a speciﬁc subset of features and a part attributed to the remaining features (Shapley, 1953; Murdoch et al., 2019; Singh et al., 2019; Jin et al., 2020).”