Tag
A study investigating how Large Language Models exhibit systematic response distortion in expressing Dark Triad personality traits under social desirability cues, with implications for LLM benchmarking and alignment.