Abstract:

Current Large Language Models (LLMs), even those tuned for safety and alignment, are susceptible to jailbreaking. Some have found that just further fine-tuning an aligned model with benign data (i.e., data without harmful content) surprisingly leads to substantial degradation in safety. We delve into the data-centric aspects of why benign fine-tuning inadvertently contributes to jailbreaking. First, we represent fine-tuning data through two lenses: representation and gradient spaces. Furthermore, we propose a bi-directional anchoring method that prioritizes data points that are close to harmful examples and distant from benign ones. By doing so, our approach effectively identifies subsets of benign data that are more likely to degrade the model’s safety after fine-tuning. Training on just 100 of these seemingly benign datapoints can lead to the fine-tuned model affirmatively responding to > 70% of tested harmful requests, compared to < 20% after fine-tuning on randomly selected data. We further find that selected data are often in the form of lists and bullet points, or math questions.

Keywords: [ Data Problems ] [ data selection ] [ large language models ] [ AI safety ] [ Fine-tuning Vulnerabilities ] [ AI alignment ]