Behavior aware method improves social media influencer role playing
From Memory to Behavior: A Behavior-Aware Role-Playing Framework for Social Media Influencers
Computation and LanguageArtificial Intelligence
Summary
It is hard for AI to imitate how real people behave in different situations, especially on social media. The authors propose a new way to teach AI to consider the situation, a person's internal feelings, and how they act, to better mimic someone. They also create a method to check how well AI imitates by comparing it to references about the person. Their tests show this method works better than existing ones for social media replies and also for fictional characters. This suggests adding behavior details helps AI act more like real or fictional people.
What this means in practice
- •For social media managers: Generate more authentic replies on social platforms by simulating how influencers respond in varied situations using behavior-aware models.
- •For game writers: Create believable dialogue for fictional characters by applying situation-dependent behavioral strategies to role-playing agents.
Authors
Ji-Lun Peng, Yi-Zhen Zhang, Chun-Nan Chou, Yun-Nung Chen
Abstract
Large language models have shown strong potential as role-playing agents for real individuals, yet faithful impersonating remains challenging. Existing in-context learning-based methods fail to capture how individuals react under different situations. In addition, LLM-based evaluation is difficult for obscure individuals. To address these challenges, we propose Situation--Internal state--Behavior Persona method to incorporate situation-dependent behavioral strategies. We further design an evaluation protocol that provides LLM evaluators with references about the impersonated individual. We evaluate our approach on a newly constructed dataset for the task of generating replies on social media. Experimental results show that our proposed method outperforms state-of-the-art ICL-based baselines, while our evaluation protocol achieves moderate correlation with human judgment. Besides, experiments on fictional-character benchmarks demonstrate that our proposed method is applicable beyond the social media setting. These findings suggest that incorporating behavioral information broadly improves the fidelity of role-playing for real individuals on social media or fictional characters.