作者位次 #5/8ACL 2026Findings· Findings of ACL
C hild E val: WHEN LARGE LANGUAGE MODELS MEET CHILDREN ’ S PERSONALITIES
Yanyan Luo · Xue Han · Chunxu Zhao · Ruiqiao Bai · Yaxing Zhang · Qian Hu · Lijun Mei · Junlan Feng
摘要
While LLMs enable personalized chatbots, their effectiveness in child-centered personalization remains unclear, as systematic evaluation of child-specific preferences is still lacking. To address this gap, we introduce ChildEval, a benchmark for evaluating LLMs’ ability to infer and follow child-...