cs.AI updates on arXiv.org 13小时前
Rethinking Safety in LLM Fine-tuning: An Optimization Perspective
index_new5.html
../../../zaker_core/zaker_tpl_static/wap/tpl_guoji1.html

 

通过系统测试,研究发现优化选择不当而非固有权衡导致语言模型安全性问题,提出EMA动量技术,有效降低有害响应,提升模型安全性能。

arXiv:2508.12531v1 Announce Type: cross Abstract: Fine-tuning language models is commonly believed to inevitably harm their safety, i.e., refusing to respond to harmful user requests, even when using harmless datasets, thus requiring additional safety measures. We challenge this belief through systematic testing, showing that poor optimization choices, rather than inherent trade-offs, often cause safety problems, measured as harmful responses to adversarial prompts. By properly selecting key training hyper-parameters, e.g., learning rate, batch size, and gradient steps, we reduce unsafe model responses from 16\% to approximately 5\%, as measured by keyword matching, while maintaining utility performance. Based on this observation, we propose a simple exponential moving average (EMA) momentum technique in parameter space that preserves safety performance by creating a stable optimization path and retains the original pre-trained model's safety properties. Our experiments on the Llama families across multiple datasets (Dolly, Alpaca, ORCA) demonstrate that safety problems during fine-tuning can largely be avoided without specialized interventions, outperforming existing approaches that require additional safety data while offering practical guidelines for maintaining both model performance and safety during adaptation.

Fish AI Reader

Fish AI Reader

AI辅助创作,多种专业模板,深度分析,高质量内容生成。从观点提取到深度思考,FishAI为您提供全方位的创作支持。新版本引入自定义参数,让您的创作更加个性化和精准。

FishAI

FishAI

鱼阅,AI 时代的下一个智能信息助手,助你摆脱信息焦虑

联系邮箱 441953276@qq.com

相关标签

语言模型 安全性优化 参数调整
相关文章