English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We examined whether embedding human attention knowledge into saliency-basedexplainable AI (XAI) methods for computer vision models could enhance theirplausibility and faithfulness. We first developed new gradient-based XAImethods for object detection models to generate object-specific explanations byextending the current methods for image classification models. Interestingly,while these gradient-based methods worked well for explaining imageclassification models, when being used for explaining object detection models,the resulting saliency maps generally had lower faithfulness than humanattention maps when performing the same task. We then developed HumanAttention-Guided XAI (HAG-XAI) to learn from human attention how to bestcombine explanatory information from the models to enhance explanationplausibility by using trainable activation functions and smoothing kernels tomaximize XAI saliency map's similarity to human attention maps. While for imageclassification models, HAG-XAI enhanced explanation plausibility at the expenseof faithfulness, for object detection models it enhanced plausibility andfaithfulness simultaneously and outperformed existing methods. The learnedfunctions were model-specific, well generalizable to other databases.

Human Attention-Guided Explainable Artificial Intelligence for Computer Vision Models