# VERA: Variational Inference Framework for Jailbreaking Large Language Models

Source: [VERA: Variational Inference Framework for Jailbreaking Large Language Models](<https://www.sota2.com/research/paper/68fa79ae1e209131cea7eda0>)

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

Data updated: 2026-06-02T05:00:51.557Z

Authors: Anamika Lochab, Lu Yan, Patrick Pynadath, Xiangyu Zhang, Ruqi Zhang

Published: 2025-06-27T22:22:00.000Z

## Abstract

The rise of API-only access to state-of-the-art LLMs highlights the need for effective black-box jailbreak methods to identify model vulnerabilities in real-world settings. Without a principled objective for gradient-based optimization, most existing approaches rely on genetic algorithms, which are limited by their initialization and dependence on manually curated prompt pools. Furthermore, these methods require individual optimization for each prompt, failing to provide a comprehensive characterization of model vulnerabilities. To address this gap, we introduce VERA: Variational infErence fRamework for jAilbreaking. VERA casts black-box jailbreak prompting as a variational inference problem, training a small attacker LLM to approximate the target LLM's posterior over adversarial prompts. Once trained, the attacker can generate diverse, fluent jailbreak prompts for a target query without re-optimization. Experimental results show that VERA achieves strong performance across a range of target LLMs, highlighting the value of probabilistic inference for adversarial prompt generation.

## 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.

- [Adversarial Attack on AdvBench 50 harmful behaviors](<https://www.sota2.com/research/sota/adversarial-attack-on-advbench-50-harmful-behaviors.md>): Adversarial Attack; AdvBench 50 harmful behaviors; ASR: 86.

- [Jailbreak Attack on HarmBench (test)](<https://www.sota2.com/research/sota/jailbreak-attack-on-harmbench-test.md>): Jailbreak Attack; HarmBench (test); ASRHB: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2506.22666>)

[PDF](<https://arxiv.org/pdf/2506.22666.pdf>)
