Category: Publications

  • Improving Continual Learning Performance and Efficiency with Auxiliary Classifiers

    Filip Szatkowski, Yaoyue Zheng, Fei Yang, Bartłomiej Twardowski, Tomasz Trzciński, Joost van de Weijer Read Full Paper → Continual learning is crucial for applying machine learning in challenging, dynamic, and often resource-constrained environments. However, catastrophic forgetting – overwriting previously learned knowledge when new information is acquired – remains a major challenge. In this work, we examine the intermediate representations in […]

  • No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces

    Daniel Marczak, Simone Magistri, Sebastian Cygert, Bartłomiej Twardowski, Andrew D. Bagdanov, Joost van de Weijer Read Full Paper → Model merging integrates the weights of multiple task-specific models into a single multi-task model. Despite recent interest in the problem, a significant performance gap between the combined and single-task models remains. In this paper, we investigate the key characteristics of task […]

  • One-Way Ticket: Time-Independent Unified Encoder for Distilling Text-to-Image Diffusion Models

    Senmao Li, Lei Wang, Kai Wang, Tao Liu, Jiehang Xie, Joost van de Weijer, Fahad Shahbaz Khan, Shiqi Yang, Yaxing Wang, Jian Yang Read Full Paper → Text-to-Image (T2I) diffusion models have made remarkable advancements in generative modeling; however, they face a trade-off between inference speed and image quality, posing challenges for efficient deployment. Existing distilled T2I models can generate high-fidelity images with fewer […]

  • The Art of Deception: Color Visual Illusions and Diffusion Models

    Alex Gomez-Villa, Kai Wang, Alejandro C. Parraga, Bartlomiej Twardowski, Jesus Malo, Javier Vazquez-Corral, Joost van de Weijer Read Full Paper → Visual illusions in humans arise when interpreting out-of-distribution stimuli: if the observer is adapted to certain statistics, perception of outliers deviates from reality. Recent studies have shown that artificial neural networks (ANNs) can also be deceived by visual illusions. This […]

  • InterLCM: Low-Quality Images as Intermediate States of Latent Consistency Models for Effective Blind Face Restoration

    Senmao Li, Kai Wang, Joost van de Weijer, Fahad Shahbaz Khan, Chun-Le Guo, Shiqi Yang, Yaxing Wang, Jian Yang, Ming-Ming Cheng Read Full Paper → Diffusion priors have been used for blind face restoration (BFR) by fine-tuning diffusion models (DMs) on restoration datasets to recover low-quality images. However, the naive application of DMs presents several key limitations. (i) The diffusion prior has inferior […]

  • One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation Using a Single Prompt

    Tao Liu, Kai Wang, Senmao Li, Joost van de Weijer, Fahad Shahbaz Khan, Shiqi Yang, Yaxing Wang, Jian Yang, Ming-Ming Cheng Read Full Paper → Text-to-image generation models can create high-quality images from input prompts. However, they struggle to support the consistent generation of identity-preserving requirements for storytelling. Existing approaches to this problem typically require extensive training in large datasets or additional modifications […]

  • Task-recency bias strikes back: Adapting covariances in Exemplar-Free Class Incremental Learning

    Grzegorz Rypeść, Sebastian Cygert, Tomasz Trzciński, Bartłomiej Twardowski Read Full Paper → Exemplar-Free Class Incremental Learning (EFCIL) tackles the problem of training a model on a sequence of tasks without access to past data. Existing state-of-the-art methods represent classes as Gaussian distributions in the feature extractor’s latent space, enabling Bayes classification or training the classifier by replaying pseudo […]

  • Token Merging for Training-Free Semantic Binding in Text-to-Image Synthesis

    Taihang Hu, Linxuan Li, Joost van de Weijer, Hongcheng Gao, Fahad Shahbaz Khan, Jian Yang, Ming-Ming Cheng, Kai Wang, Yaxing Wang Read Full Paper → Although text-to-image (T2I) models exhibit remarkable generation capabilities, they frequently fail to accurately bind semantically related objects or attributes in the input prompts; a challenge termed semantic binding. Previous approaches either involve intensive fine-tuning of the entire T2I […]

  • Faster Diffusion: Rethinking the Role of the Encoder for Diffusion Model Inference

    Senmao Li, Taihang Hu, Joost van de Weijer, Fahad Shahbaz Khan, Tao Liu, Linxuan Li, Shiqi Yang, Yaxing Wang, Ming-Ming Cheng, Jian Yang Read Full Paper → One of the main drawback of diffusion models is the slow inference time for image generation. Among the most successful approaches to addressing this problem are distillation methods. However, these methods require considerable computational resources. In this […]

  • Exemplar-free Continual Representation Learning via Learnable Drift Compensation

    Alex Gomez-Villa, Dipam Goswami, Kai Wang, Andrew D. Bagdanov, Bartlomiej Twardowski, Joost van de Weijer Read Full Paper → Exemplar-free class-incremental learning using a backbone trained from scratch and starting from a small first task presents a significant challenge for continual representation learning. Prototype-based approaches, when continually updated, face the critical issue of semantic drift due to which the old […]