SeqLoRA: Continual Multi-Concept Generation Without Catastrophic Forgetting

arXiv·medium signal

Addresses the open problem of sequentially fine-tuning text-to-image diffusion models on multiple custom concepts without forgetting earlier ones. Uses bilevel orthogonal LoRA adapters that maintain per-concept fidelity while preventing interference across concepts. Practical for anyone building personalized image generation pipelines where new concepts arrive over time.