7 papers
UnHype: CLIP-Guided Hypernetworks for Dynamic LoRA Unlearning
Piotr Wójcik, Maksym Petrenko, Wojciech Gromski +2
Recent advances in large-scale diffusion models have intensified concerns about their potential misuse, particularly in generating realistic yet harmful or socially disruptive cont…
From Unlearning to UNBRANDING: A Benchmark for Trademark-Safe Text-to-Image Generation
Dawid Malarz, Filip Manjak, Maciej ZiÄba +2
The rapid progress of text-to-image diffusion models raises significant concerns regarding the unauthorized reproduction of trademarked content. While prior work targets general co…
GaINeR: Geometry-Aware Implicit Network Representation
Weronika Jakubowska, MikoÅaj ZieliÅski, RafaÅ Tobiasz +4
Implicit Neural Representations (INRs) are widely used for modeling continuous 2D images, enabling high-fidelity reconstruction, super-resolution, and compression. Architectures su…
Spiking World Model with Multi-Compartment Neurons for Model-based Reinforcement Learning
Yinqian Sun, Feifei Zhao, Mingyang Lv +1
Brain-inspired spiking neural networks (SNNs) have garnered significant research attention in algorithm design and perception applications. However, their potential in the decision…
Classifier-free Guidance with Adaptive Scaling
Dawid Malarz, Artur Kasymov, Maciej ZiÄba +2
Classifier-free guidance (CFG) is an essential mechanism in contemporary text-driven diffusion models. In practice, in controlling the impact of guidance we can see the trade-off b…
Neural Surface Priors for Editable Gaussian Splatting
Jakub Szymkowiak, Weronika Jakubowska, Dawid Malarz +5
In computer graphics and vision, recovering easily modifiable scene appearance from image data is crucial for applications such as content creation. We introduce a novel method tha…