163 citations · 1.4k across the 139 of their papers we have counts for
12 papers · 2 filters
Identifying and Mitigating Model Failures through Few-shot CLIP-aided Diffusion Generation
Atoosa Chegini, Soheil Feizi
Deep learning models can encounter unexpected failures, especially when dealing with challenging sub-populations. One common reason for these failures is the occurrence of objects…
Instruct2Attack: Language-Guided Semantic Adversarial Attacks
Jiang Liu, Chen Wei, Yuxiang Guo +5
We propose Instruct2Attack (I2A), a language-guided semantic attack that generates semantically meaningful perturbations according to free-form language instructions. We make use o…
Exploring Geometry of Blind Spots in Vision Models
Sriram Balasubramanian, Gaurang Sriramanan, Vinu Sankar Sadasivan +1
Despite the remarkable success of deep neural networks in a myriad of settings, several works have demonstrated their overwhelming sensitivity to near-imperceptible perturbations,…
Localizing and Editing Knowledge in Text-to-Image Generative Models
Samyadeep Basu, Nanxuan Zhao, Vlad Morariu +2
Text-to-Image Diffusion Models such as Stable-Diffusion and Imagen have achieved unprecedented quality of photorealism with state-of-the-art FID scores on MS-COCO and other generat…
EditVal: Benchmarking Diffusion Based Text-Guided Image Editing Methods
Samyadeep Basu, Mehrdad Saberi, Shweta Bhardwaj +5
A plethora of text-guided image editing methods have recently been developed by leveraging the impressive capabilities of large-scale diffusion-based generative models such as Imag…
PRIME: Prioritizing Interpretability in Failure Mode Extraction
Keivan Rezaei, Mehrdad Saberi, Mazda Moayeri +1
In this work, we study the challenge of providing human-understandable descriptions for failure modes in trained image classification models. Existing works address this problem by…
Robustness of AI-Image Detectors: Fundamental Limits and Practical Attacks
Mehrdad Saberi, Vinu Sankar Sadasivan, Keivan Rezaei +4
In light of recent advancements in generative AI models, it has become essential to distinguish genuine content from AI-generated one to prevent the malicious usage of fake materia…
Adapting Self-Supervised Representations to Multi-Domain Setups
Neha Kalibhat, Sam Sharpe, Jeremy Goodsitt +2
Current state-of-the-art self-supervised approaches, are effective when trained on individual domains but show limited generalization on unseen domains. We observe that these model…
Identifying Interpretable Subspaces in Image Representations
Neha Kalibhat, Shweta Bhardwaj, Bayan Bruss +3
We propose Automatic Feature Explanation using Contrasting Concepts (FALCON), an interpretability framework to explain features of image representations. For a target feature, FALC…
Distilling Knowledge from Text-to-Image Generative Models Improves Visio-Linguistic Reasoning in CLIP
Samyadeep Basu, Shell Xu Hu, Maziar Sanjabi +2
Image-text contrastive models like CLIP have wide applications in zero-shot classification, image-text retrieval, and transfer learning. However, they often struggle on composition…
Text-To-Concept (and Back) via Cross-Model Alignment
Mazda Moayeri, Keivan Rezaei, Maziar Sanjabi +1
We observe that the mapping between an image's representation in one model to its representation in another can be learned surprisingly well with just a linear layer, even across d…
Strong Baselines for Parameter Efficient Few-Shot Fine-tuning
Samyadeep Basu, Daniela Massiceti, Shell Xu Hu +1
Few-shot classification (FSC) entails learning novel classes given only a few examples per class after a pre-training (or meta-training) phase on a set of base classes. Recent work…