11 papers
Representation Recycling for Streaming Video Analysis
Can Ufuk Ertenli, Ramazan Gokberk Cinbis, Emre Akbas
We present StreamDEQ, a method that aims to infer frame-wise representations on videos with minimal per-frame computation. Conventional deep networks perform feature extraction fro…
Caption Bottleneck Models
Seref Baris Cagliyan, Umut Ozdemir, Merve Tapli +1
Concept Bottleneck Models (CBMs) provide interpretability by routing predictions through a layer of human-understandable concepts. However, defining an optimal concept set for a sp…
Explaining CLIP Zero-shot Predictions Through Concepts
Onat Ozdemir, Anders Christensen, Stephan Alaniz +2
Large-scale vision-language models such as CLIP have achieved remarkable success in zero-shot image recognition, yet their predictions remain largely opaque to human understanding.…
Rethinking Concept Bottleneck Models: From Pitfalls to Solutions
Merve Tapli, Quentin Bouniot, Wolfgang Stammer +2
Concept Bottleneck Models (CBMs) ground predictions in human-understandable concepts but face fundamental limitations: the absence of a metric to pre-evaluate concept relevance, th…
Intrinsic Dimensionality as a Model-Free Measure of Class Imbalance
ÃaÄrı Eser, Zeynep Sonat Baltacı, Emre AkbaÅ +1
Imbalance in classification tasks is commonly quantified by the cardinalities of examples across classes. This, however, disregards the presence of redundant examples and inherent…
Unsupervised Image Classification with Adaptive Nearest Neighbor Selection and Cluster Ensembles
Melih Baydar, Emre Akbas
Unsupervised image classification, or image clustering, aims to group unlabeled images into semantically meaningful categories. Early methods integrated representation learning and…