Related papers: Linear Adversarial Concept Erasure
Models trained on real-world data often mirror and exacerbate existing social biases. Traditional methods for mitigating these biases typically require prior knowledge of the specific biases to be addressed, such as gender or racial biases,…
Concept erasure has emerged as a promising technique for mitigating the risk of harmful content generation in diffusion models by selectively unlearning undesirable concepts. The common principle of previous works to remove a specific…
Concept erasure techniques for text-to-video (T2V) diffusion models report substantial suppression of sensitive content, yet current evaluation is limited to checking whether the target concept is absent from generated frames, treating…
Adversarial training can be used to learn models that are robust against perturbations. For linear models, it can be formulated as a convex optimization problem. Compared to methods proposed in the context of deep learning, leveraging the…
Advanced text-to-image diffusion models raise safety concerns regarding identity privacy violation, copyright infringement, and Not Safe For Work content generation. Towards this, unlearning methods have been developed to erase these…
Over the last years, various sentence embedders have been an integral part in the success of current machine learning approaches to Natural Language Processing (NLP). Unfortunately, multiple sources have shown that the bias, inherent in the…
Solving large-scale optimization on-the-fly is often a difficult task for real-time computer graphics applications. To tackle this challenge, model reduction is a well-adopted technique. Despite its usefulness, model reduction often…
Transformer language models (LMs) have been shown to represent concepts as directions in the latent space of hidden activations. However, for any human-interpretable concept, how can we find its direction in the latent space? We present a…
Modern machine learning datasets can have biases for certain representations that are leveraged by algorithms to achieve high performance without learning to solve the underlying task. This problem is referred to as "representation bias".…
Interpretability benefits the theoretical understanding of representations. Existing word embeddings are generally dense representations. Hence, the meaning of latent dimensions is difficult to interpret. This makes word embeddings like a…
Machine unlearning for text-to-image diffusion models aims to selectively remove undesirable concepts from pre-trained models without costly retraining. Current unlearning methods share a common weakness: erased concepts return when the…
Concept erasure, which fine-tunes diffusion models to remove undesired or harmful visual concepts, has become a mainstream approach to mitigating unsafe or illegal image generation in text-to-image models.However, existing removal methods…
Large pre-trained language models are often trained on large volumes of internet data, some of which may contain toxic or abusive language. Consequently, language models encode toxic information, which makes the real-world usage of these…
The joint optimization of the reconstruction and classification error is a hard non convex problem, especially when a non linear mapping is utilized. In order to overcome this obstacle, a novel optimization strategy is proposed, in which a…
Generative autoencoders offer a promising approach for controllable text generation by leveraging their latent sentence representations. However, current models struggle to maintain coherent latent spaces required to perform meaningful text…
Neural networks are among the state-of-the-art techniques for language modeling. Existing neural language models typically map discrete words to distributed, dense vector representations. After information processing of the preceding…
Word embeddings derived from human-generated corpora inherit strong gender bias which can be further amplified by downstream models. Some commonly adopted debiasing approaches, including the seminal Hard Debias algorithm, apply…
Recently, contrastive learning has been shown to be effective in improving pre-trained language models (PLM) to derive high-quality sentence representations. It aims to pull close positive examples to enhance the alignment while push apart…
Text-to-image diffusion models have been demonstrated with undesired generation due to unfiltered large-scale training data, such as sexual images and copyrights, necessitating the erasure of undesired concepts. Most existing methods focus…
Deep learning models are vulnerable to adversarial examples and make incomprehensible mistakes, which puts a threat on their real-world deployment. Combined with the idea of adversarial training, preprocessing-based defenses are popular and…