PsyMem: Fine-grained psychological alignment and Explicit Memory Control for Advanced Role-Playing LLMs

TL;DR

PsyMem introduces a framework for role-playing LLMs that uses detailed psychological attributes and explicit memory control to improve character consistency and reliability in social simulation tasks.

Contribution

It presents a novel approach combining fine-grained psychological indicators with explicit memory alignment training for more reliable role-playing in LLMs.

Findings

PsyMem-Qwen outperforms baseline models in human-likeness.

Achieves higher character fidelity in role-playing.

Demonstrates effective dynamic memory-controlled responses.

Abstract

Existing LLM-based role-playing methods often rely on superficial textual descriptions or simplistic metrics, inadequately modeling both intrinsic and extrinsic character dimensions. Additionally, they typically simulate character memory with implicit model knowledge or basic retrieval augment generation without explicit memory alignment, compromising memory consistency. The two issues weaken reliability of role-playing LLMs in several applications, such as trustworthy social simulation. To address these limitations, we propose PsyMem, a novel framework integrating fine-grained psychological attributes and explicit memory control for role-playing. PsyMem supplements textual descriptions with 26 psychological indicators to detailed model character. Additionally, PsyMem implements memory alignment training, explicitly trains the model to align character's response with memory, thereby enabling dynamic memory-controlled responding during inference. By training Qwen2.5-7B-Instruct on our specially designed dataset (including 5,414 characters and 38,962 dialogues extracted from novels), the resulting model, termed as PsyMem-Qwen, outperforms baseline models in role-playing, achieving the best performance in human-likeness and character fidelity.