# ToxSearch: Evolving Prompts for Toxicity Search in Large Language Models

arXiv: https://arxiv.org/abs/2511.12487 · HTML: https://mlsanity.com/paper/2511.12487/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 16 of its 22 references that are in the map.
- First posted on arXiv 16 Nov 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Safety alignment and jailbreak attacks](https://mlsanity.com/alignment/safety-alignment-and-jailbreak-attacks/)
- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 16 of 16)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models](https://mlsanity.com/paper/2310.04451/) (2023)
- [GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts](https://mlsanity.com/paper/2309.10253/) (2023)
- [Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution](https://mlsanity.com/paper/2309.16797/) (2023)
- [Generating Natural Language Adversarial Examples](https://mlsanity.com/paper/1804.07998/) (2018)
- [BAE: BERT-based Adversarial Examples for Text Classification](https://mlsanity.com/paper/2004.01970/) (2020)
- [Red-Teaming Large Language Models using Chain of Utterances for Safety-Alignment](https://mlsanity.com/paper/2308.09662/) (2023)
- [Rainbow Teaming: Open-Ended Generation of Diverse Adversarial Prompts](https://mlsanity.com/paper/2402.16822/) (2024)
- [EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers](https://mlsanity.com/paper/2309.08532/) (2023)
- [Language Model Crossover: Variation through Few-Shot Prompting](https://mlsanity.com/paper/2302.12170/) (2023)
- [Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic](https://mlsanity.com/paper/2402.11746/) (2024)
- [InstOptima: Evolutionary Multi-objective Instruction Optimization via Large Language Model-based Instruction Operators](https://mlsanity.com/paper/2310.17630/) (2023)
- [No Offense Taken: Eliciting Offensiveness from Language Models](https://mlsanity.com/paper/2310.00892/) (2023)
- [How Toxic Can You Get? Search-based Toxicity Testing for Large Language Models](https://mlsanity.com/paper/2501.01741/) (2025)

## Abstract (arXiv)

Large Language Models remain vulnerable to adversarial prompts that elicit toxic content even after safety alignment. We present ToxSearch, a black-box evolutionary framework that tests model safety by evolving prompts in a synchronous steady-state loop. The system employs a diverse set of operators, including lexical substitutions, negation, back-translation, paraphrasing, and two semantic crossover operators, while a moderation oracle provides fitness guidance. Operator-level analysis shows heterogeneous behavior: lexical substitutions offer the best yield-variance trade-off, semantic-similarity crossover acts as a precise low-throughput inserter, and global rewrites exhibit high variance with elevated refusal costs. Using elite prompts evolved on LLaMA 3.1 8B, we observe practically meaningful but attenuated cross-model transfer, with toxicity roughly halving on most targets, smaller LLaMA 3.2 variants showing the strongest resistance, and some cross-architecture models retaining higher toxicity. These results suggest that small, controllable perturbations are effective vehicles for systematic red-teaming and that defenses should anticipate cross-model reuse of adversarial prompts rather than focusing only on single-model hardening.