Red-Teaming Large Language Models using Chain of Utterances for Safety-Alignment

by   Rishabh Bhardwaj, et al.
Singapore University of Technology and Design

Larger language models (LLMs) have taken the world by storm with their massive multi-tasking capabilities simply by optimizing over a next-word prediction objective. With the emergence of their properties and encoded knowledge, the risk of LLMs producing harmful outputs increases, making them unfit for scalable deployment for the public. In this work, we propose a new safety evaluation benchmark RED-EVAL that carries out red-teaming. We show that even widely deployed models are susceptible to the Chain of Utterances-based (CoU) prompting, jailbreaking closed source LLM-based systems such as GPT-4 and ChatGPT to unethically respond to more than 65 also demonstrate the consistency of the RED-EVAL across 8 open-source LLMs in generating harmful responses in more than 86 Next, we propose RED-INSTRUCT–An approach for the safety alignment of LLMs. It constitutes two phases: 1) HARMFULQA data collection: Leveraging CoU prompting, we collect a dataset that consists of 1.9K harmful questions covering a wide range of topics, 9.5K safe and 7.3K harmful conversations from ChatGPT; 2) SAFE-ALIGN: We demonstrate how the conversational dataset can be used for the safety alignment of LLMs by minimizing the negative log-likelihood over helpful responses and penalizing over harmful responses by gradient accent over sample loss. Our model STARLING, a fine-tuned Vicuna-7B, is observed to be more safely aligned when evaluated on RED-EVAL and HHH benchmarks while preserving the utility of the baseline models (TruthfulQA, MMLU, and BBH).


Llama 2: Open Foundation and Fine-Tuned Chat Models

In this work, we develop and release Llama 2, a collection of pretrained...

Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned

We describe our early efforts to red team language models in order to si...

The Poison of Alignment

From the perspective of content safety issues, alignment has shown to li...

Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts

Text-to-image diffusion models, e.g. Stable Diffusion (SD), lately have ...

Jailbroken: How Does LLM Safety Training Fail?

Large language models trained for safety and harmlessness remain suscept...

XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models

Without proper safeguards, large language models will readily follow mal...

Please sign up or login with your details

Forgot password? Click here to reset