作者位次 #4/10ACL 2026Poster· Natural Language Processing
Beyond Noise: Characterizing Creative Potential in Unverifiable LLM Hallucinations
Yu Yan · Chunhong Zhang · Haiyu Zhao · Ziyang Zeng · Zihao Liu · Yongkang Wu · Jianzhou Diao · YiJie Chen · Shujie Wang · Zheng Hu
摘要
In knowledge-intensive creative tasks, Large Language Models (LLMs) often generate outputs that extend beyond established knowledge, making direct verification against current evidence impractical. Unlike factual hallucinations checked against ground truth, such outputs arise naturally in creativ...