Generative image steganography aims to conceal secret information in generated images without arousing suspicion. However, in practical scenarios involving high-capacity embedding or lossy transmission, existing methods still suffer from limited extraction accuracy. The main challenge lies in accurately recovering the secret-embedded latent vectors from stego images. To address this issue, we propose CIF, a constrained inversion framework designed to achieve accurate message extraction. Specifically, CIF reduces dynamic structural errors by enforcing linear consistency in the latent space, meanwhile reduces numerical integration errors by adaptively adjusting the integration order according to local trajectory stability. Experimental results show that our method reduces latent reconstruction error by more than 35\% and achieves higher message extraction accuracy than existing approaches.
@article{arxiv.2508.00434,
title = {CIF: A Constrained Inversion Framework for Reliable Message Extraction in Diffusion-Based Generative Steganography},
author = {Yuqi Qian and Yun Cao and Meiyang Lv and Haocheng Fu},
journal= {arXiv preprint arXiv:2508.00434},
year = {2025}
}