Efficiency vs. Alignment: Investigating Safety and Fairness Risks in Parameter-Efficient Fine-Tuning of LLMs

Title:Efficiency vs. Alignment: Investigating Safety and Fairness Risks in Parameter-Efficient Fine-Tuning of LLMs

Abstract:Organizations increasingly adapt Large Language Models (LLMs) from public repositories such as HuggingFace to downstream tasks. Prior work shows that even fine-tuning on benign datasets can weaken safety alignment, raising a practical question: does benign parameter-efficient fine-tuning (PEFT) also affect safety and fairness? We present the first large-scale, systematic study showing that benign PEFT can significantly alter both. We fine-tune four instruction-tuned model families (Meta-Llama-3-8B, Qwen2.5-7B, Mistral-7B, and Gemma-7B) with four widely used PEFT methods: LoRA, IA3, Prompt-Tuning, and P-Tuning. In total, we evaluate 235 conversationally fine-tuned variants across eleven safety hazard categories and nine fairness dimensions. We assess generalization beyond conversational tuning by incorporating a compact extension focused on coding tasks, involving 96 additional fine-tuned models. Results show that benign PEFT can induce detrimental alignment shifts. Adapter-based methods (LoRA, IA3) are generally safer and less disruptive to fairness, whereas prompt-based methods more often reduce safety and worsen fairness accuracy. Base model choice strongly moderates these effects: LLaMA is comparatively stable, Qwen shows modest gains, Gemma exhibits the steepest safety decline, and Mistral is the most variable. The coding-task extension also produces alignment shifts relative to base models, but matched comparisons with the conversational task reveal limited task-level differences. Overall, safety improvements do not reliably transfer to fairness, and no single configuration optimizes every fairness metric. These findings support a practical guideline for safety-critical deployment: benign intent does not guarantee safe behaviour; start from a well-aligned base model, favour adapter-based PEFT, and audit safety and fairness at the category level.

Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

arXiv:2511.00382 [cs.AI]

(or arXiv:2511.00382v2 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2511.00382 Focus to learn more arXiv-issued DOI via DataCite

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.