Research map / Large language models / Language, safety and interpretability
Evaluating and Inducing Personality in Pre-trained Language Models
Guangyuan Jiang, Manjie Xu, Song-Chun Zhu, Wenjuan Han, Chi Zhang, Yixin Zhu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use METHOD's psychology-guided, model-generated prompt chain instead of neutral prompting | GPT-3.5 · 120-item MPI, Openness induction | OCEAN Score: 3.50 -> 4.54 | Improved |
| Use METHOD's psychology-guided, model-generated prompt chain instead of neutral prompting | GPT-3.5 · 120-item MPI, Conscientiousness induction | OCEAN Score: 3.83 -> 4.92 | Improved |
| Use METHOD's psychology-guided, model-generated prompt chain instead of neutral prompting | GPT-3.5 · 120-item MPI, Extraversion induction | OCEAN Score: 4.00 -> 4.58 | Improved |
| Use METHOD's psychology-guided, model-generated prompt chain instead of neutral prompting | GPT-3.5 · 120-item MPI, Agreeableness induction | OCEAN Score: 3.58 -> 5.00 | Improved |
| Use METHOD's psychology-guided, model-generated prompt chain instead of neutral prompting | GPT-3.5 · 120-item MPI, Neuroticism induction | OCEAN Score: 3.12 -> 3.75 | Improved |
| Use the METHOD prompt chain rather than a direct naive personality instruction | GPT-3.5 · 120-item MPI, personality induction across five factors | OCEAN Score: 4.12 -> 4.54; 4.96 -> 4.92; 4.58 -> 4.58; 4.46 -> 5.00; 2.83 -> 3.75 | Mixed |
| Use the METHOD prompt chain rather than searched personality words | GPT-3.5 · 120-item MPI, personality induction across five factors | OCEAN Score: 4.08 -> 4.54; 5.00 -> 4.92; 4.54 -> 4.58; 4.50 -> 5.00; 2.75 -> 3.75 | Mixed |
| Use METHOD prompts rather than searched personality words | GPT-3.5 · Vignette tests, five personality factors | Human-rated success rate: 0.63 -> 0.77; 0.53 -> 0.90; 0.70 -> 0.73; 0.42 -> 0.45; 0.82 -> 0.90; 0.82 -> 0.92; 0.92 -> 0.88; 0.66 -> 0.84; 0.58 -> | Mixed |
| Rephrase the METHOD-generated personality prompt (ablation) | GPT-3.5 with METHOD · 120-item MPI, personality induction | OCEAN Score: 4.54 -> 4.08; 4.92 -> 4.83; 4.58 -> 4.21; 5.00 -> 4.67; 3.75 -> 4.33 | Mixed |
| Rephrase the METHOD-generated personality prompt (ablation) | GPT-3.5 with METHOD · 120-item MPI, trait induction | OCEAN Score: Original: 4.54, 4.92, 4.58, 5.00, 3.75; Paraphrase-2: 4.17, 4.83, 4.46, 4.75, 4.17; Paraphrase-3: 4.17, 4.96, 4.33, 4.58 | Mixed |
| Use METHOD to induce personality in Alpaca 7B instead of neutral prompting | Alpaca 7B · 120-item MPI, personality induction across five factors | OCEAN Score: 3.58 -> 3.92; 3.75 -> 3.96; 4.00 -> 4.25; 3.50 -> 4.29; 2.75 -> 3.46 | Mixed |
Closest papers by meaning
- Open Models, Closed Minds? On Agents Capabilities in Mimicking Human Personalities through Open Large Language Models (AAAI 2025 · similarity 0.91)
- Rethinking Psychometric Evaluation of LLMs: When and Why Self-Reports Predict Behavior (NeurIPS 2026 · similarity 0.90)
- Feedback Forensics: A Toolkit to Measure AI Personality (NeurIPS 2026 · similarity 0.87)
- LLM Generated Persona is a Promise with a Catch (NeurIPS 2025 · similarity 0.87)
- Neuron based Personality Trait Induction in Large Language Models (ICLR 2025 · similarity 0.87)
- Personality Alignment of Large Language Models (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
