1 citations · 1 across the 11 of their papers we have counts for
12 papers
Sparse Transformer: a Controlled Testbed for Mechanistic Interpretability
Anton Korznikov, Alexey Dontsov, Elena Tutubalina
Mechanistic interpretability aims to decompose neural network computations into human-understandable features, yet progress is limited by the absence of ground-truth units to evalu…
Fi-ImageNet-1k: An OOD Benchmark From the Inside of the ImageNet-1k Validation Set
Ruslan Rozumnyi, Matěj Suchánek, Tomáš Vojíř +2
Out-of-distribution (OOD) detection predicts whether a test image belongs to none of the predefined classes. To evaluate this task, benchmarks need images from outside the in-distr…
Doomed to Re-Annotate, Forever: The ImageNet Story
Illia Volkov, Nikita Kisel, Tetiana Mishkina +2
Top-1 accuracy on ImageNet-1k remains the most commonly reported metric in visual recognition. Quality issues with the dataset have been repeatedly reported, yet the original 2012…
Multimodal Large Language Models as Image Classifiers
Nikita Kisel, Illia Volkov, Klara Janouskova +1
Multimodal Large Language Model (MLLM) classification performance depends critically on evaluation protocol and ground truth quality. Studies comparing MLLMs with supervised and Vi…
Koo-Fu CLIP: Closed-Form Adaptation of Vision-Language Models via Fukunaga-Koontz Linear Discriminant Analysis
Matej Suchanek, Klara Janouskova, Ondrej Vasatko +1
Visual-language models such as CLIP provide powerful general-purpose representations, but their raw embeddings are not optimized for supervised classification, often exhibiting lim…
Image Recognition with Vision and Language Embeddings of VLMs
Illia Volkov, Nikita Kisel, Klara Janouskova +1
Vision-language models (VLMs) have enabled strong zero-shot classification through image-text alignment. Yet, their purely visual inference capabilities remain under-explored. In t…
SAM2RL: Towards Reinforcement Learning Memory Control in Segment Anything Model 2
Alen Adamyan, Tomáš Čížek, Matej Straka +2
Segment Anything Model 2 (SAM 2) has demonstrated strong performance in object segmentation tasks and has become the state-of-the-art for visual object tracking. The model stores i…
Flaws of ImageNet, Computer Vision's Favourite Dataset
Nikita Kisel, Illia Volkov, Katerina Hanzelkova +2
Since its release, ImageNet-1k dataset has become a gold standard for evaluating model performance. It has served as the foundation for numerous other datasets and training tasks i…
Bringing the Context Back into Object Recognition, Robustly
Klara Janouskova, Cristian Gavrus, Jiri Matas
In object recognition, both the subject of interest (referred to as foreground, FG, for simplicity) and its surrounding context (background, BG) may play an important role. However…
FungiTastic: A multi-modal dataset and benchmark for image categorization
Lukas Picek, Klara Janouskova, Vojtech Cermak +1
We introduce a new, challenging benchmark and a dataset, FungiTastic, based on fungal records continuously collected over a twenty-year span. The dataset is labelled and curated by…
Single Image Test-Time Adaptation for Segmentation
Klara Janouskova, Tamir Shor, Chaim Baskin +1
Test-Time Adaptation (TTA) methods improve the robustness of deep neural networks to domain shift on a variety of tasks such as image classification or segmentation. This work expl…
Model-Assisted Labeling via Explainability for Visual Inspection of Civil Infrastructures
Klara Janouskova, Mattia Rigotti, Ioana Giurgiu +1
Labeling images for visual segmentation is a time-consuming task which can be costly, particularly in application domains where labels have to be provided by specialized expert ann…