AI RESEARCH
Value Entanglement: Conflation Between Different Kinds of Good In (Some) Large Language Models
arXiv CS.AI
•
ArXi:2602.19101v2 Announce Type: replace-cross Value alignment of Large Language Models (LLMs) requires us to empirically measure these models' actual, acquired representation of value. Among the characteristics of value representation in humans is that they distinguish among value of different kinds. We investigate whether LLMs likewise distinguish three different kinds of good: moral, grammatical, and economic.