DICE: Discrete Inversion Enabling Controllable Editing for Multinomial Diffusion and Masked Generative Models

Discrete diffusion models have achieved success in tasks like image generation and masked language modeling but face limitations in controlled content editing. We introduce DICE (Discrete Inversion for Controllable Editing), the first approach to enable precise inversion for discrete diffusion models, including multinomial diffusion and masked generative models. By recording noise sequences and masking patterns during the reverse diffusion process, DICE enables accurate reconstruction and flexible editing of discrete data without the need for predefined masks or attention manipulation. We demonstrate the effectiveness of DICE across both image and text domains, evaluating it on models such as VQ-Diffusion, Paella, and RoBERTa. Our results show that DICE preserves high data fidelity while enhancing editing capabilities, offering new opportunities for fine-grained content manipulation in discrete spaces.

@article{he2025_2410.08207,
  title={ DICE: Discrete Inversion Enabling Controllable Editing for Multinomial Diffusion and Masked Generative Models },
  author={ Xiaoxiao He and Ligong Han and Quan Dao and Song Wen and Minhao Bai and Di Liu and Han Zhang and Martin Renqiang Min and Felix Juefei-Xu and Chaowei Tan and Bo Liu and Kang Li and Hongdong Li and Junzhou Huang and Faez Ahmed and Akash Srivastava and Dimitris Metaxas },
  journal={arXiv preprint arXiv:2410.08207},
  year={ 2025 }
}