Fame Fades, Nature Remains: Disentangling the Character Identity of Role-Playing Agents
This work addresses the challenge of improving character fidelity in RPAs for applications like gaming and interactive AI, though it is incremental as it builds on existing LLM-based RPAs.
The paper tackled the problem of weakly formalized character identity in Role-Playing Agents (RPAs) by proposing a multidimensional construct that disentangles characters into Parametric and Attributive Identity layers, revealing that famous characters lose their initial advantage over turns and that negative social natures are a primary bottleneck in RPA fidelity.
Despite the rapid proliferation of Role-Playing Agents (RPAs) based on Large Language Models (LLMs), the structural dimensions defining a character's identity remain weakly formalized, often treating characters as arbitrary text inputs. In this paper, we propose the concept of \textbf{Character Identity}, a multidimensional construct that disentangles a character into two distinct layers: \textbf{(1) Parametric Identity}, referring to character-specific knowledge encoded from the LLM's pre-training, and \textbf{(2) Attributive Identity}, capturing fine-grained behavioral properties such as personality traits and moral values. To systematically investigate these layers, we construct a unified character profile schema and generate both Famous and Synthetic characters under identical structural constraints. Our evaluation across single-turn and multi-turn interactions reveals two critical phenomena. First, we identify \textit{"Fame Fades"}: while famous characters hold a significant advantage in initial turns due to parametric knowledge, this edge rapidly vanishes as models prioritize accumulating conversational context over pre-trained priors. Second, we find that \textit{"Nature Remains"}: while models robustly portray general personality traits regardless of polarity, RPA performance is highly sensitive to the valence of morality and interpersonal relationships. Our findings pinpoint negative social natures as the primary bottleneck in RPA fidelity, guiding future character construction and evaluation.