The field of generative AI is currently seeing a surge in text-to-image generation. Among open-source projects, Stable Diffusion stands out as the state-of-the-art. Artists and service providers often customize diffusion models for unique textures. However, there is a lack of privacy protection for users' input text prompts, output images, and customized models on servers. Ensuring privacy is essential for user trust and safeguarding intellectual property. Current privacy-preserving diffusion models rely on fully homomorphic encryption (FHE), which is time-intensive and can compromise image quality. We introduce Anonymous-Diffusion, a diffusion as a service (DAAS) framework. This framework maintains privacy without using FHE by exploiting the irreversible nature of neural network layers and the characteristic that predicted noise in the diffusion process follows a normalized Gaussian distribution. Furthermore, we ensure anonymity by Smart Contract and Blockchain. User can use this service on demand anonymously. In comparison to existing research like HE-diffusion, which incurs a 200% time overhead and noticeable quality degradation, our protocol achieves the same level of security with only a 4% time overhead and no loss in image quality. To our knowledge, this is the first solution to achieve these results without FHE while preserving high-quality image output. © 2025 IEEE.

Cross-chain communication is the cryptographic technology that sends and receives tokens or data among multiple blockchains. A parent-child blockchain system is one of the methods of cross-chain communication. Existing studies apply Byzantine fault-tolerant (BFT) mechanisms to achieve consensus among multiple N blockchains. However, cross-chain communication based on PBFT requires O(N2) communication complexity to achieve consensus mechanisms among multiple N blockchains (no scalability). In addition, existing cross-chain communication do not propose a scheme to prevent a malicious act by users who mediate tokens (no security). Furthermore, there is no existing scheme that simultaneously satisfies unforgeability, which prevents the sender of a token from forging the sent token, and unlinkability, which protects the privacy of the sender and receiver (no unlinkability and no unforgeability). In this paper, we propose a Privacy-Preserving Scalable Cross-Chain Communication (PPSCCC) among multiple blockchains that solve the above problems. In PPSCCC, by applying the commitment scheme to the token mediators, the communication complexity among blockchains for consensus mechanism can be achieved with O(N). We also prove unlinkability and unforgeability by using the hiding and binding properties of the commitment scheme. In addition, by introducing Proof of Commitment Reputation (PoCR), we achieve a scheme to prevent malicious behavior of users who mediate tokens. We then implement our proposed scheme. Finally, we compare our scheme with existing schemes to highlight the strengths of our scheme and show that our scheme is more practical. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

Text-to-image generation is trending in the generative AI field. Stable Diffusion is the state-of-the-art among open-source projects. Many artists and service providers customize the diffusion model for special textures. However, there is no protection for the privacy of the user's input text prompt, output image, and the customized model on the server. Privacy is crucial for user trust and protecting intellectual property. Existing privacy-preserving diffusion models use fully homomorphic encryption (FHE), which is time-consuming and can degrade image quality. We propose Privacy-Diffusion, a framework that preserves privacy without FHE by leveraging the irreversible properties of neural network layers and the property that in the diffusion process, the predicted noise is a normalized Gaussian distribution. Our framework protects clients' input text prompts and generated images from the server and safeguards customized models from clients. Compared with existing research HE-diffusion which spent 200% extra time and visible quality loss, our protocol can reach the same security level with only 4% extra time and has no quality loss. To our knowledge, we are the first to achieve this goal without FHE while maintaining high-quality image output. © 2025 IEEE.

Cross-chain communication is a cryptographic technology that allows communication between different blockchains. Cross-chain communication is a technology that is gaining attention to expand the potential of blockchains, but it is difficult to construct securely and efficiently. The construction of the M+1st-price sealed bid auction in cross-chain communication is also one of the difficult schemes to realize. The existing M+1st-price sealed bid auction scheme uses a commitment scheme that needs to open the winner's bid amount after the auction is finished, which does not protect privacy. Furthermore, it is inefficient to construct it using the existing commitment scheme since there are multiple blockchain participants as bidders in cross-chain communication. In this paper, we propose a privacy-preserving and efficient M+1st-price sealed bid auction in cross-chain communication. To achieve privacy-preserving, we propose a scheme that can determine the maximum bid amount without disclosing the largest bidder using a vector commitment scheme. Furthermore, the vector commitment scheme allows the verifier to aggregate multiple bid values into one commitment. Thus, efficiency can be achieved by aggregating the multiplier commitment value by using a vector commitment scheme. Also, by combining the vector commitment scheme and encryption scheme, only the verifier can compare the bid value. Finally, we compare our scheme and other existing schemes to emphasize the contribution of our scheme. © 2024 IEEE.

As Blockchain grows in size, the resources to vali-date that a transaction is true increase. Key-value commitments are suggested to solve this problem. If we use Key-Value Commitments, we can authenticate affiliation by three elements, commitment, key-value pair, and proof. But in this method, if users are offline, which means they cannot receive updated information, their proof will expire. In this research, we propose a method that users to update proof with the information they return online. The previous scheme allows updating the proof in O (1), but n users must execute the update. Furthermore, adding an off-line update function requires O(T) data storage. In our proposal, we need only O(log n) update proofs, and we can realize it only one user to update the proof. © 2024 IEEE.