Personality shapes memory use in dialogue agents for consistent character
PersMem: Internalizing Personality into Dual-Pathway Memory for LLM Agents
Artificial Intelligence
Summary
Many dialogue agents use a fixed personality but handle their memories separately, which can make their reactions feel inconsistent. The authors propose PersMem, a system that makes an agent’s memory processing depend on its personality. By linking personality to how emotions are noted, memories are kept, and relevant memories are recalled, the agent behaves more consistently with its assigned character. This also allows people to check if the agent stays true to its personality during conversations.
What this means in practice
- •For chatbot developers: Create chatbots with personality-consistent memory retrieval to improve user experience and character believability in conversations.
- •For game designers: Enhance non-player characters by linking their memory use to personality traits, improving narrative engagement and believable character behavior.
Authors
Hanzhong Zhang, Ziwei Xiang, Weicheng Xie, Shizhe Liu, Siyang Song
Abstract
The profile of a role-playing agent usually depends on the pre-defined personality in a system prompt, whereas its memory processing pipeline, including prioritisation of stored memories and subsequent retrieval, remains independent of this personality. This separation causes the agent's memory processing to be inconsistent with the pre-defined personality, and makes it difficult to validate whether agent behaviours follow this personality. In this paper, we propose Personality-Integrated Memory (PersMem), which integrates personality into the agent's memory processing pipeline, making it consistently personality-dependent. PersMem processes memory using four steps, where the personality is mapped to operation-specific parameters controlling: (i) affective appraisal annotating emotion states of the user input; (ii) retention of previously stored memories along with the current input; (iii) passive affect-driven memory retrieval exploring memories similar to user input in semantics and personality-guided emotions; and (iv) active goal-driven memory retrieval that refines and selects passively retrieved memories for the reply. Consequently, consistency with the pre-defined personality can be examined by inspecting memory-processing traces during human-agent interactions. We evaluate these personality-dependent differences in attachment and Big Five settings. PersMem exceeds the chance baseline for four-way attachment classification by 23.1 percentage points. In Big Five dialogue comparisons, PersMem achieves 67.5% accuracy, 6.7 percentage points above a baseline using uniformly sampled memories. On CoSER, PersMem achieves an average score of 66.13, with scores of 69.33 for Character Fidelity and 84.33 for Storyline Quality. Together, these results show that PersMem produces distinguishable personality-related memory-processing patterns.