1
PICACO: Pluralistic In-Context Value Alignment of LLMs via Total Correlation Optimization
提出PICACO方法，利用总相关优化实现大模型的多元上下文价值对齐，减少有害输出同时适配多元偏好。
arXiv:2507.16679v3 Announce Type: replace-cross Abstract: In-Context Learning has shown great potential for aligning Large Language Models (LLMs) with…