Related papers: Probing Classifiers are Unreliable for Concept Rem…

Neural networks employ spurious correlations in their predictions, resulting in decreased performance when these correlations do not hold. Recent works suggest fixing pretrained representations and training a classification head that does…

Saliency post-hoc explainability methods are important tools for understanding increasingly complex NLP models. While these methods can reflect the model's reasoning, they may not align with human intuition, making the explanations not…

Deep neural networks and other intricate Artificial Intelligence (AI) models have reached high levels of accuracy on many biomedical natural language processing tasks. However, their applicability in real-world use cases may be limited due…

Numerous approaches have been recently proposed for learning fair representations that mitigate unfair outcomes in prediction tasks. A key motivation for these methods is that the representations can be used by third parties with unknown…

Concept-based machine learning methods have increasingly gained importance due to the growing interest in making neural networks interpretable. However, concept annotations are generally challenging to obtain, making it crucial to leverage…

It has been demonstrated that deep neural networks are prone to noisy examples particular adversarial samples during inference process. The gap between robust deep learning systems in real world applications and vulnerable neural networks…

Extracting fashion attributes from images of people wearing clothing/fashion accessories is a very hard multi-class classification problem. Most often, even catalogues of fashion do not have all the fine-grained attributes tagged due to…

Machine learning systems based on deep neural networks, being able to produce state-of-the-art results on various perception tasks, have gained mainstream adoption in many applications. However, they are shown to be vulnerable to…

As pretrained models are increasingly shared on the web, ensuring that models can forget or delete sensitive, copyrighted, or private information upon request has become crucial. Machine unlearning has been proposed to address this…

Current approaches for explaining deep learning systems applied to musical data provide results in a low-level feature space, e.g., by highlighting potentially relevant time-frequency bins in a spectrogram or time-pitch bins in a piano…

Prototype learning, a popular machine learning method designed for inherently interpretable decisions, leverages similarities to learned prototypes for classifying new data. While it is mainly applied in computer vision, in this work, we…

Post-hoc importance attribution methods are a popular tool for "explaining" Deep Neural Networks (DNNs) and are inherently based on the assumption that the explanations can be applied independently of how the models were trained.…

Recently, with the advancement of deep learning, several applications in text classification have advanced significantly. However, this improvement comes with a cost because deep learning is vulnerable to adversarial examples. This weakness…

The rise of social networks has not only facilitated communication but also allowed the spread of harmful content. Although significant advances have been made in detecting toxic language in textual data, the exploration of concept-based…

Ante-hoc interpretability methods based on prototypes provide highly accurate explanations by utilizing the intuitive "this looks like that" reasoning paradigm. On the other hand, post-hoc models can explain predictions for a single image…

Neural network models have a reputation for being black boxes. We propose to monitor the features at every layer of a model and measure how suitable they are for classification. We use linear classifiers, which we refer to as "probes",…

Recent work has demonstrated the successful extraction of training data from generative language models. However, it is not evident whether such extraction is feasible in text classification models since the training objective is to predict…

Identification of input data points relevant for the classifier (i.e. serve as the support vector) has recently spurred the interest of researchers for both interpretability as well as dataset debugging. This paper presents an in-depth…

Text guided diffusion models are used by millions of users, but can be easily exploited to produce harmful content. Concept unlearning methods aim at reducing the models' likelihood of generating harmful content. Traditionally, this has…

Machine Learning models face increased concerns regarding the storage of personal user data and adverse impacts of corrupted data like backdoors or systematic bias. Machine Unlearning can address these by allowing post-hoc deletion of…