cs.LGOct 31, 2025

Diffusion LLMs are Natural Adversaries for any LLM

Authors: David Lüdke, Tom Wollschläger, Paul Ungermann, Stephan Günnemann, Leo Schwinn

Organizations: Department of Computer Science & Munich Data Science Institute Technical University of Munich, Germany

Abstract

We introduce a novel framework that transforms the resource-intensive (adversarial) prompt optimization problem into an efficient, amortized inference task. Our core insight is that pretrained, non-autoregressive generative LLMs, such as Diffusion LLMs, which model the joint distribution over prompt-response pairs, can serve as powerful surrogates for prompt search. This approach enables direct conditional generation of prompts, effectively replacing costly discrete optimization with a small number of parallelizable samples. We provide a probabilistic analysis demonstrating that under mild fidelity assumptions, only a few conditional samples are required to recover high-reward (harmful) prompts. Empirically, our method substantially outperforms prior attacks in both success rate and compute cost, producing low-perplexity, diverse jailbreaks that transfer to a wide range of black-box target models, including robustly trained and proprietary LLMs. Beyond adversarial prompting, our framework opens new directions for red teaming, automated prompt optimization, and leveraging emerging Flow- and Diffusion-based LLMs.

Figures & tables

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Automated jailbreak attack targeting multiple defense strategies

    Jun 15, 2026Qi Wang, Chengcheng Wan, Weijia He +4Adversarial Prompt GenerationLLM Jailbreak Attacks

  2. Adaptive Prompt Embedding Optimization for LLM Jailbreaking

    Apr 27, 2026Miles Q. Li, Benjamin C. M. Fung, Boyang Li +2Adversarial Prompt GenerationAdversarial Attacks on LLMs

  3. JailbreakOPT: Iterative Jailbreak Prompt Optimization with Macro-Action

    Jun 9, 2026Ge Shi, Jun Yin, Donglin Xie +3Adversarial Prompt GenerationLarge Language Model-Guided Optimization