Defining Cost Function of Steganography with Large Language Models

2025-12-10

AI Summary

This work addresses the longstanding challenge in steganography where cost functions rely either on expert domain knowledge or large-scale labeled datasets. We propose the first large language model (LLM)-driven automated framework for cost function construction. Our method employs a two-stage paradigm: (1) LLM-guided program synthesis to generate executable initial cost functions, and (2) iterative refinement via evolutionary search, guided by detection accuracy feedback from steganalyzers. Crucially, this approach eliminates dependence on human priors or extensive annotated data, enabling end-to-end evaluability, retrainability, and evolutionary discovery of cost functions. Experiments demonstrate that the automatically generated cost functions significantly enhance steganographic security—reducing false positive rates by over 30% across mainstream steganalyzers. Our framework establishes a novel, fully automated pipeline for adaptive, data-efficient steganographic design.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Large Multimodal Models (LMMs)
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

In this paper, we make the first attempt towards defining cost function of steganography with large language models (LLMs), which is totally different from previous works that rely heavily on expert knowledge or require large-scale datasets for cost learning. To achieve this goal, a two-stage strategy combining LLM-guided program synthesis with evolutionary search is applied in the proposed method. In the first stage, a certain number of cost functions in the form of computer program are synthesized from LLM responses to structured prompts. These cost functions are then evaluated with pretrained steganalysis models so that candidate cost functions suited to steganography can be collected. In the second stage, by retraining a steganalysis model for each candidate cost function, the optimal cost function(s) can be determined according to the detection accuracy. This two-stage strategy is performed by an iterative fashion so that the best cost function can be collected at the last iteration. Experiments show that the proposed method enables LLMs to design new cost functions of steganography that significantly outperform existing works in terms of resisting steganalysis tools, which verifies the superiority of the proposed method. To the best knowledge of the authors, this is the first work applying LLMs to the design of advanced cost function of steganography, which presents a novel perspective for steganography design and may shed light on further research.

Problem

• Defines steganography cost functions using large language models
• Applies LLM-guided program synthesis and evolutionary search
• Enhances steganography resistance against detection tools

Innovation

• Using LLMs to generate steganography cost functions via program synthesis
• Evaluating cost functions with pretrained steganalysis models for selection
• Iteratively refining cost functions through evolutionary search and retraining