1
Coherence Maximization Improves Pluralistic Alignment
这篇论文提出通过最大化一致性来改善多元对齐，为AI安全与价值对齐研究提供了新思路。
arXiv:2606.03110v1 Announce Type: new Abstract: Aligning AI systems with diverse human values requires value specifications grounded in concrete examp…