As LLM-based agents are increasingly used in long-term interactions, cumulative memory is critical for enabling personalization and maintaining stylistic consistency. However, most existing systems adopt an ``all-or-nothing'' approach to memory usage: incorporating all relevant past information can lead to \textit{Memory Anchoring}, where the agent is trapped by past interactions, while excluding memory entirely results in under-utilization and the loss of important interaction history. We show that an agent's reliance on memory can be modeled as an explicit and user-controllable dimension. We first introduce a behavioral metric of memory dependence to quantify the influence of past interactions on current outputs. We then propose \textbf{Stee}rable \textbf{M}emory Agent, \texttt{SteeM}, a framework that allows users to dynamically regulate memory reliance, ranging from a fresh-start mode that promotes innovation to a high-fidelity mode that closely follows interaction history. Experiments across different scenarios demonstrate that our approach consistently outperforms conventional prompting and rigid memory masking strategies, yielding a more nuanced and effective control for personalized human-agent collaboration.
@article{arxiv.2601.05107,
title = {Controllable Memory Usage: Balancing Anchoring and Innovation in Long-Term Human-Agent Interaction},
author = {Muzhao Tian and Zisu Huang and Xiaohua Wang and Jingwen Xu and Zhengkang Guo and Qi Qian and Yuanzhe Shen and Kaitao Song and Jiakang Yuan and Changze Lv and Xiaoqing Zheng},
journal= {arXiv preprint arXiv:2601.05107},
year = {2026}
}