paper-with-me

홈 › Papers

Human Interpretation of Saliency-based Explanation Over Text

2022-01-27 · Hendrik Schuff, Alon Jacovi, Heike Adel, Yoav Goldberg, Ngoc Thang Vu

While a lot of research in explainable AI focuses on producing effective explanations, less work is devoted to the question of how people understand and interpret the explanation. In this work, we focus on this question through a study of saliency-based explanations over textual data. Feature-attribution explanations of text models aim to communicate which parts of the input text were more influential than others towards the model decision. Many current explanation methods, such as gradient-based or Shapley value-based methods, provide measures of importance which are well-understood mathematically. But how does a person receiving the explanation (the explainee) comprehend it? And does their understanding match what the explanation attempted to communicate? We empirically investigate the effect of various factors of the input, the feature-attribution explanation, and visualization procedure, on laypeople's interpretation of the explanation. We query crowdworkers for their interpretation on tasks in English and German, and fit a GAMM model to their responses considering the factors of interest. We find that people often mis-interpret the explanations: superficial and unrelated factors, such as word length, influence the explainees' importance assignment despite the explanation communicating importance directly. We then show that some of this distortion can be attenuated: we propose a method to adjust saliencies based on model estimates of over- and under-perception, and explore bar charts as an alternative to heatmap saliency visualization. We find that both approaches can attenuate the distorting effect of specific factors, leading to better-calibrated understanding of the explanation.

📄 PDF Abstract BibTeX arXiv:2201.11569

Code (1)

boschresearch/human-interpretation-saliency 공식 구현

Methods 이 논문이 사용한 방법론

Heatmap 설명 없음

Similar Papers 제목 키워드 기반

Neighboring Words Affect Human Interpretation of Saliency Explanations

2023-05-04 · Alon Jacovi, Hendrik Schuff, Heike Adel, Ngoc Thang Vu 외

Word-level saliency explanations ("heat maps over words") are often used to communicate feature-attribution in text-based models. Recent studies found that superficial factors such as word length can distort human interp…

Deriving Explanation of Deep Visual Saliency Models

2021-09-08 · Sai Phani Kumar Malladi, Jayanta Mukhopadhyay, Chaker Larabi, Santanu Chaudhury

Deep neural networks have shown their profound impact on achieving human level performance in visual saliency prediction. However, it is still unclear how they learn the task and what it means in terms of understanding h…

Explainable ModelsSaliency Prediction

Rethinking Saliency Maps: A Cognitive Human Aligned Taxonomy and Evaluation Framework for Explanations

2025-11-17 · Yehonatan Elisha, Seffi Cohen, Oren Barkan, Noam Koenigstein arxiv

Saliency maps are widely used for visual explanations in deep learning, but a fundamental lack of consensus persists regarding their intended purpose and alignment with diverse user queries. This ambiguity hinders the ef…

Saliency Map Verbalization: Comparing Feature Importance Representations from Model-free and Instruction-based Methods

2022-10-13 · Nils Feldhus, Leonhard Hennig, Maximilian Dustin Nasert, Christopher Ebert 외

Saliency maps can explain a neural model's predictions by identifying important input features. They are difficult to interpret for laypeople, especially for instances with many features. In order to make them more acces…

Abstractive Text SummarizationFeature ImportanceSentiment Analysistext-classification+2

MASE: Interpretable NLP Models via Model-Agnostic Saliency Estimation

2025-12-04 · Zhou Yang, Shunyan Luo, Jiazhen Zhu, Fang Jin arxiv

Deep neural networks (DNNs) have made significant strides in Natural Language Processing (NLP), yet their interpretability remains elusive, particularly when evaluating their intricate decision-making processes. Traditio…