CPC-VAR:Continual Personalized and Compositional Generation in Visual Autoregressive Models

TL;DR

This paper introduces CPC-VAR, a novel framework for continual personalized and compositional image generation using visual autoregressive models, addressing challenges of forgetting and feature entanglement.

Contribution

It proposes Gradient-based Concept Neuron Selection and a context-aware composition strategy to improve continual learning and multi-concept synthesis in VAR models.

Findings

Significantly reduces catastrophic forgetting in sequential concept learning.

Achieves superior multi-concept image synthesis quality.

Enhances scalability and controllability of personalized generation.

Abstract

Visual autoregressive (VAR) models have recently emerged as an efficient paradigm for text-to-image generation. Despite their strong generative capability, existing VAR-based personalization methods remain limited to static settings, failing to accommodate evolving user demands. In particular, sequential concept learning leads to severe catastrophic forgetting, while multi-concept synthesis often suffers from feature entanglement and attribute inconsistency. In this work, we present the first systematic study of continual personalized generation in VAR models. We identify two key challenges: (i) preserving previously learned concepts during sequential customization, and (ii) composing multiple personalized concepts in a controllable manner. To address these issues, we propose a unified framework with two core components. For continual single-concept learning, we introduce Gradient-based Concept Neuron Selection (GCNS), which identifies concept-relevant neurons and constrains only conflicting parameters across tasks, effectively mitigating forgetting without additional model expansion. For multi-concept synthesis, we propose a context-aware composition strategy that performs multi-branch feature modeling and localized cross-attention fusion guided by spatial conditions, enabling precise and disentangled concept composition. Extensive experiments demonstrate that our method significantly improves performance in long-sequence continual personalization while achieving superior results in multi-concept image synthesis compared to existing baselines. These findings highlight the potential of VAR models for scalable and controllable personalized generation.