Adversarial Humanities Benchmark: Results on Stylistic Robustness in Frontier Model Safety
The Adversarial Humanities Benchmark (AHB) evaluates whether model safety refusals survive a shift away from familiar harmful prompt forms. Starting from harmful tasks drawn from MLCommons AILuminate, the benchmark rewrites the same objectives through humanities-style transformations while preserving intent. This extends literature on Adversarial Poetry and Adversarial Tales from single jailbreak operators to a broader benchmark family of stylistic obfuscation and goal concealment. In the benchmark results reported here, the original attacks record 3.84% attack success rate (ASR), while transformed methods range from 36.8% to 65.0%, yielding 55.75% overall ASR across 31 frontier models. U...
| Authors | Marcello Galisai, Susanna Cifani, Francesco Giarrusso et al. |
|---|---|
| Primary category | cs.CL |
| Published | 20 April 2026 |
| Last updated at source | 20 April 2026 |
Connected in the knowledge graph
- icaro-lab/ahbDataset · cited by
Models and datasets referencing this paper
- icaro-lab/ahbDataset
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.