Path attribution methods are a popular tool to interpret visual model's prediction on an input. They integrate model gradients for the input features over path defined between and reference, thereby satisfying certain desirable theoretical properties. However, their reliability hinges choice of reference. Moreover, they do not exhibit weak dependence input, which leads counter-intuitive feature...

Perceptual representations pick out individuals and attribute properties to them. This paper considers the role of perceptual attribution in determining or guiding reference objects. We consider three extant models relation between reference–all guides reference, no a privileged subset attributions reference–and argue that empirical evidence undermines all three. then defend flexible-attributiv...

Abstract Academic attribution, the direct acknowledgement of external sources, is investigated in two corpora novice academic English, representing first and second language writing linguistics. The forms uses attribution are analysed a formal-functional framework. There an overall underrepresentation learner corpus. However, have similar proportional distribution integral non-integral but diff...