Improving the Distributional Alignment of LLMs using Supervision

ArXi:2507.00439v4 Announce Type: replace The ability to accurately align LLMs with diverse population groups on subjective questions would have great value. In this work, we show that adding simple supervision can consistently improve the alignment of LLM-generated distributions with diverse population groups, as measured across three datasets spanning public health, public opinion, and values and beliefs. Beyond evaluating average alignment, we also report how alignment varies across specific groups.