Unified Multimodal Generative Models (UMGMs) unify visual understanding and image generation within a single autoregressive framework. However, their ability to continually learn new tasks is severely hindered by catastrophic forgetting, both within a modality (intra-modal) and across modalities (inter-modal). While intra-modal forgetting has been studied in prior continual learning (CL) work, inter-modal forgetting remains largely unexplored. In this paper, we identify and empirically validate this phenomenon in UMGMs and provide a theoretical explanation rooted in gradient conflict between modalities. To address both intra- and inter-modal forgetting, we propose Modality-Decoupled Experts (MoDE), a lightweight and scalable architecture that isolates modality-specific updates to mitigate the gradient conflict and leverages knowledge distillation to prevent catastrophic forgetting and preserve pre-trained capabilities. Unlike previous CL methods that remain modality-coupled and suffer from modality gradient conflict, MoDE explicitly decouples modalities to prevent interference. Experiments across diverse benchmarks demonstrate that MoDE significantly mitigates both inter- and intra-modal forgetting, outperforming prior CL baselines in unified multimodal generation settings.

Catastrophic forgetting is a significant challenge in online continual learning (OCL), especially for nonstationary data streams that do not have well-defined task boundaries. This challenge is exacerbated by the memory constraints and privacy concerns inherent in rehearsal buffers. To tackle catastrophic forgetting, in this paper, we introduce Online-LoRA, a novel framework for task-free OCL. Online-LoRA allows to finetune pre-trained Vision Transformer (ViT) models in real-time to address the limitations of rehearsal buffers and leverage pre-trained models' performance benefits. As the main contribution, our approach features a novel online weight regularization strategy to identify and consolidate important model parameters. Moreover, Online-LoRA leverages the training dynamics of loss values to enable the automatic recognition of the data distribution shifts. Extensive experiments across many task-free OCL scenarios and benchmark datasets (including CIFAR-100, ImageNet-R, ImageNet-S, CUB-200 and CORe50) demonstrate that Online-LoRA can be robustly adapted to various ViT architectures, while achieving better performance compared to SOTA methods11Code: https://github.com/Christina200/online-LoRA-official.git.

Great progress has been made in StyleGAN-based image editing. To associate with preset attributes, most existing approaches focus on supervised learning for semantically meaningful latent space traversal directions, and each manipulation step is typically determined for an individual attribute. To address this limitation, we propose a Text-guided Unsupervised StyleGAN Latent Transformation (TUSLT) model, which adaptively infers a single transformation step in the latent space of StyleGAN to simultaneously manipulate multiple attributes on a given input image. Specifically, we adopt a two-stage architecture for a latent mapping network to break down the transformation process into two manageable steps. Our network first learns a diverse set of semantic directions tailored to an input image, and later nonlinearly fuses the ones associated with the target attributes to infer a residual vector. The resulting tightly interlinked two-stage architecture delivers the flexibility to handle diverse attribute combinations. By leveraging the cross-modal text-image representation of CLIP, we can perform pseudo annotations based on the semantic similarity between preset attribute text descriptions and training images, and further jointly train an auxiliary attribute classifier with the latent mapping network to provide semantic guidance. We perform extensive experiments to demonstrate that the adopted strategies contribute to the superior performance of TUSLT.

Data interpolation is typically used to explore and understand the latent representation learnt by a deep network. Naive linear interpolation may induce mismatch between the interpolated data and the underlying manifold of the original data. In this paper, we propose an Adversarial Adaptive Interpolation (AdvAI) approach for facilitating representation learning and image synthesis in autoencoders. To determine an interpolation path that stays on the manifold, we incorporate an interpolation correction module, which learns to offset the deviation from the manifold. Further, we perform matching with a prior distribution to control the characteristics of the representation. The data synthesized from random codes along with interpolation-based regularization are in turn used to constrain the representation learning process. In the experiments, the superior performance of the proposed approach demonstrates the effectiveness of AdvAI and associated regularizers in a variety of downstream tasks.

Generative adversarial network (GAN)-based models possess superior capability of high-fidelity image synthesis. There are a wide range of semantically meaningful directions in the latent representation space of well-trained GANs, and the corresponding latent space walks are meaningful for semantic controllability in the synthesized images. To explore the underlying organization of a latent space, we propose an unsupervised Density-Preserving Latent Semantics Exploration model (DP-LaSE). The important latent directions are determined by maximizing the variations in intermediate features, while the correlation between the directions is minimized. Considering that latent codes are sampled from a prior distribution, we adopt a density-preserving regularization approach to ensure latent space walks are maintained in iso-density regions, since moving to a higher/lower density region tends to cause unexpected transformations. To further refine semantics-specific transformations, we perform subspace learning over intermediate feature channels, such that the transformations are limited to the most relevant subspaces. Extensive experiments on a variety of benchmark datasets demonstrate that DP-LaSE is able to discover interpretable latent space walks, and specific properties of synthesized images can thus be precisely controlled.