A Novel Differential Feature Learning for Effective Hallucination Detection and Classification

Wenkai Wang, Vincent Lee, Yizhen Zheng

Large language model hallucination represents a critical challenge where outputs deviate from factual accuracy due to distributional biases in training data. While recent investiga…