{"ID":2922176,"CreatedAt":"2026-06-02T02:42:49.606572591Z","UpdatedAt":"2026-06-02T19:07:03.846039099Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.00860","arxiv_id":"2606.00860","title":"GenPT: Beyond Self-Report for Reliable LLM Psychometrics via Generative Projective Testing","abstract":"Self-report questionnaires remain the prevailing tool for probing the psychological states of persona-conditioned agents (PC-Agents). However, classical instruments inherit two well-known threats: contamination from training corpora and directional bias driven by social-desirability or contextual framing. To overcome these methodological bottlenecks, we ask whether projective paradigms can be adapted into a robust psychometric tool. We introduce \\textbf{GenPT} (Generative Projective Testing), which reformulates TAT, Rorschach, and SCT with newly generated stimuli and organizes assessment as a three-stage pipeline to derive standardized psychological indicators and target states. Evaluating PC-Agents induced via CharacterRAG and AnnaAgent profiles, we benchmark GenPT's reliability and validity against classical questionnaires. The results indicate that questionnaires exhibit systematic directional shifts under social-desirability framing, most strongly on suicide ideation. In contrast, GenPT's collected behavioral patterns stay near the symmetric baseline. Furthermore, under a longitudinal counselling context, GenPT-based depression assessment shifts by roughly an order of magnitude more than the questionnaire counterpart when Qwen3 serves as the backbone. Overall, GenPT complements self-report methods in scenarios where contamination resistance, bias asymmetry, and context sensitivity matter. Code and stimuli can be found at https://github.com/sci-m-wang/GenPT.","short_abstract":"Self-report questionnaires remain the prevailing tool for probing the psychological states of persona-conditioned agents (PC-Agents). However, classical instruments inherit two well-known threats: contamination from training corpora and directional bias driven by social-desirability or contextual framing. To overcome t...","url_abs":"https://arxiv.org/abs/2606.00860","url_pdf":"https://arxiv.org/pdf/2606.00860v1","authors":"[\"Ming Wang\",\"Shuang Wu\",\"Bixuan Wang\",\"Lu Lin\",\"Yuxin Chen\",\"Xiaocui Yang\",\"Daling Wang\",\"Shi Feng\",\"Yifei Zhang\",\"Yufan Sun\"]","published":"2026-05-30T19:20:26Z","proceeding":"cs.SI","tasks":"[\"cs.SI\",\"cs.AI\",\"cs.CL\"]","methods":"[\"Large Language Model\",\"Generative Adversarial Network\"]","has_code":false,"code_links":[{"ID":612648,"CreatedAt":"2026-06-02T02:42:49.606572591Z","UpdatedAt":"2026-06-02T02:42:49.606572591Z","DeletedAt":null,"paper_id":2922176,"paper_url":"https://arxiv.org/abs/2606.00860","paper_title":"GenPT: Beyond Self-Report for Reliable LLM Psychometrics via Generative Projective Testing","repo_url":"https://github.com/sci-m-wang/GenPT","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}