# PRoADS: Provably Secure and Robust Audio Diffusion Steganography with latent optimization and backward Euler Inversion

Source: [PRoADS: Provably Secure and Robust Audio Diffusion Steganography with latent optimization and backward Euler Inversion](<https://www.sota2.com/research/paper/69b24876d030084c22e2f837>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-13T05:00:04.336Z

Authors: YongPeng Yan, Yanan Li, Qiyang Xiao, Yanzhen Ren

Published: 2026-03-11T01:29:30.000Z

## Abstract

This paper proposes PRoADS, a provably secure and robust audio steganographic framework based on audio diffusion models. As a generative steganography scheme, PRoADS embeds secret messages into the initial noise of diffusion models via orthogonal matrix projection. To address the reconstruction errors in diffusion inversion that cause high bit error rates (BER), we introduce Latent Optimization and Backward Euler Inversion to minimize the latent reconstruction and diffusion inversion errors. Comprehensive experiments demonstrate that our scheme sustains a remarkably low BER of 0.15\\% under 64 kbps MP3 compression, significantly outperforming existing methods and exhibiting strong robustness.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Audio Steganography on AudioCaps](<https://www.sota2.com/research/sota/audio-steganography-on-audiocaps.md>): Audio Steganography; AudioCaps; BER (Original): 0.09.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.10314>)

[PDF](<https://arxiv.org/pdf/2603.10314.pdf>)