可控記憶使用：在長期人機互動中平衡錨定效應與創新性

摘要

隨著基於大型語言模型的智慧體日益廣泛應用於長期互動場景，累積記憶能力對於實現個人化服務與維持風格一致性至關重要。然而，現有系統大多採用「全有或全無」的記憶使用模式：若完整納入過往互動資訊，可能導致「記憶錨定」現象，使智慧體受困於歷史對話軌跡；反之若完全排除記憶，則會造成重要互動歷史的未充分運用與遺失。我們的研究表明，智慧體對記憶的依賴程度可被建模為一個可顯式調控的維度。我們首先提出記憶依賴性的行為度量指標，用以量化過往互動對當前輸出的影響程度。接著推出可調控記憶智慧體框架SteeM，該框架允許使用者動態調節記憶依賴強度，範圍涵蓋激發創新的「全新啟動」模式，到嚴格遵循互動歷史的「高保真」模式。跨場景實驗表明，我們的方法在個性化人機協作中，能持續優於傳統提示法與剛性記憶遮蔽策略，實現更細膩且高效的控制效果。

English

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 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 Steerable Memory Agent, 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.

可控記憶使用：在長期人機互動中平衡錨定效應與創新性

Controllable Memory Usage: Balancing Anchoring and Innovation in Long-Term Human-Agent Interaction

摘要

Support