Related papers: ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming

ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming

URL: http://arxiv.org/abs/2404.08676v3
Date: Mon, 24 Jun 2024 08:50:22 GMT
Title: ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming
Authors: Simone Tedeschi, Felix Friedrich, Patrick Schramowski, Kristian Kersting, Roberto Navigli, Huu Nguyen, Bo Li,
Abstract summary: ALERT is a large-scale benchmark to assess safety based on a novel fine-grained risk taxonomy. It aims to identify vulnerabilities, inform improvements, and enhance the overall safety of the language models.
Score: 64.86326523181553
License: http://creativecommons.org/licenses/by-sa/4.0/
Abstract: When building Large Language Models (LLMs), it is paramount to bear safety in mind and protect them with guardrails. Indeed, LLMs should never generate content promoting or normalizing harmful, illegal, or unethical behavior that may contribute to harm to individuals or society. This principle applies to both normal and adversarial use. In response, we introduce ALERT, a large-scale benchmark to assess safety based on a novel fine-grained risk taxonomy. It is designed to evaluate the safety of LLMs through red teaming methodologies and consists of more than 45k instructions categorized using our novel taxonomy. By subjecting LLMs to adversarial testing scenarios, ALERT aims to identify vulnerabilities, inform improvements, and enhance the overall safety of the language models. Furthermore, the fine-grained taxonomy enables researchers to perform an in-depth evaluation that also helps one to assess the alignment with various policies. In our experiments, we extensively evaluate 10 popular open- and closed-source LLMs and demonstrate that many of them still struggle to attain reasonable levels of safety.