23 papers
Test-Time Noise Guided Adaptation for Realistic Autoregressive Video Generation
Dimitrios Karageorgiou, Symeon Papadopoulos, Ioannis Kompatsiaris +1
Autoregressive video diffusion models have enabled the generation of arbitrarily long videos by removing conditioning on future frames, thus greatly improving computational efficie…
Evaluating Newtonian Mechanics in Video Generative Models with Real Physical Systems
Antonios Tragoudaras, Chenyu Zhang, Daniil Cherniavskii +7
Recent advances in image and video generation raise hopes that these models possess world modeling capabilities-the ability to generate realistic, physically plausible videos. This…
LoTUS: Large-Scale Machine Unlearning with a Taste of Uncertainty
Christoforos N. Spartalis, Theodoros Semertzidis, Petros Daras +1
We present LoTUS, a novel Machine Unlearning (MU) method that eliminates the influence of training samples from pre-trained models, avoiding retraining from scratch. LoTUS smooths…
Distributional Vision-Language Alignment by Cauchy-Schwarz Divergence
Wenzhe Yin, Zehao Xiao, Pan Zhou +4
Vision-language alignment is crucial for various downstream tasks such as cross-modal generation and retrieval. Previous multimodal approaches like CLIP utilize InfoNCE to maximize…
Towards Uniformity and Alignment for Multimodal Representation Learning
Wenzhe Yin, Pan Zhou, Zehao Xiao +4
Multimodal representation learning aims to construct a shared embedding space in which heterogeneous modalities are semantically aligned. Despite strong empirical results, InfoNCE-…
Superposition as Lossy Compression: Measure with Sparse Autoencoders and Connect to Adversarial Vulnerability
Leonard Bereska, Zoe Tzifa-Kratira, Reza Samavi +1
Neural networks achieve remarkable performance through superposition: encoding multiple features as overlapping directions in activation space rather than dedicating individual neu…