MoKus: Leveraging Cross-Modal Knowledge Transfer for Knowledge-Aware Concept Customization

Abstract

Concept customization typically binds rare tokens to a target concept. Unfortunately, these approaches often suffer from unstable performance as the pretraining data seldom contains these rare tokens. Meanwhile, these rare tokens fail to convey the inherent knowledge of the target concept. Consequently, we introduce Knowledge-aware Concept Customization, a novel task aiming at binding diverse textual knowledge to target visual concepts. This task requires the model to identify the knowledge within the text prompt to perform high-fidelity customized generation. Meanwhile, the model should efficiently bind all the textual knowledge to the target concept. Therefore, we propose MoKus, a novel framework for knowledge-aware concept customization. Our framework relies on a key observation: cross-modal knowledge transfer, where modifying knowledge within the text modality naturally transfers to the visual modality during generation. Inspired by this observation, MoKus contains two stages: (1) In visual concept learning, we first learn the anchor representation to store the visual information of the target concept. (2) In textual knowledge updating, we update the answer for the knowledge queries to the anchor representation, enabling high-fidelity customized generation. To further comprehensively evaluate our proposed MoKus on the new task, we introduce the first benchmark for knowledge-aware concept customization: KnowCusBench. Extensive evaluations have demonstrated that MoKus outperforms state-of-the-art methods. Moreover, the cross-model knowledge transfer allows MoKus to be easily extended to other knowledge-aware applications like virtual concept creation and concept erasure. We also demonstrate the capability of our method to achieve improvements on world knowledge benchmarks.

Figures (10)

• Figure 1: Comparison between previous concept customization and knowledge-aware concept customization.(a) Previous customization techniques use rare tokens (e.g., <sks>) to represent a target concept. However, these tokens lack clear semantic meaning, which can cause unstable generation results. Furthermore, rare tokens cannot store knowledge about the target concept. (b) In our proposed knowledge-aware concept customization, it links the target visual concept with several pieces of textual knowledge, which enables robust and high-fidelity reconstruction and customization.
• Figure 2: Examples to show the cross-modal knowledge transfer phenomenon.
• Figure 3: Overview of our MoKus. (a) Visual Concept Learning: We bind the visual information of the target concept to the anchor representation. We achieve this connection by fine-tuning the LoRA parameters. (b) Textual Knowledge Updating: We first convert the knowledge into a query and input it into the LLM encoder. We then extract the hidden states and update directions from the updatable layers to calculate a parameter shift. Finally, we add this shift to the original parameters of these layers.
• Figure 4: Visualization of our KnowCusBench.
• Figure 5: Qualitative comparison of reconstruction. We directly use the knowledge to reconstruct the target concept.