PsychAdapter: Adapting LLM Transformers to Reflect Traits, Personality and Mental Health
This enables the creation of chatbots with specific personality profiles, clinical training tools, and machine translations tailored to psychological traits, addressing the need for personalized AI language generation in psychology and related domains.
The authors tackled the problem of language models generating 'average' language by proposing PsychAdapter, a lightweight transformer modification that uses trait-language patterns to generate text reflecting specified personality, demographic, and mental health characteristics, achieving 87.3% accuracy for Big Five personalities and 96.7% for depression and life satisfaction in expert evaluations.
Artificial intelligence-based language generators are now a part of most people's lives. However, by default, they tend to generate "average" language without reflecting the ways in which people differ. Here, we propose a lightweight modification to the standard language model transformer architecture - "PsychAdapter" - that uses empirically derived trait-language patterns to generate natural language for specified personality, demographic, and mental health characteristics (with or without prompting). We applied PsychAdapters to modify OpenAI's GPT-2, Google's Gemma, and Meta's Llama 3 and found generated text to reflect the desired traits. For example, expert raters evaluated PsychAdapter's generated text output and found it matched intended trait levels with 87.3% average accuracy for Big Five personalities, and 96.7% for depression and life satisfaction. PsychAdapter is a novel method to introduce psychological behavior patterns into language models at the foundation level, independent of prompting, by influencing every transformer layer. This approach can create chatbots with specific personality profiles, clinical training tools that mirror language associated with psychological conditionals, and machine translations that match an authors reading or education level without taking up LLM context windows. PsychAdapter also allows for the exploration psychological constructs through natural language expression, extending the natural language processing toolkit to study human psychology.