Controllable Memory Usage: Balancing Anchoring and Innovation in Long-Term Human-Agent Interaction
Abstract
A framework is presented that enables dynamic regulation of memory reliance in LLM-based agents, allowing users to control the balance between innovation and historical fidelity in long-term interactions.
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.
Community
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.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- PersonaMem-v2: Towards Personalized Intelligence via Learning Implicit User Personas and Agentic Memory (2025)
- MemBuilder: Reinforcing LLMs for Long-Term Memory Construction via Attributed Dense Rewards (2026)
- Learning User Preferences Through Interaction for Long-Term Collaboration (2026)
- Mem-PAL: Towards Memory-based Personalized Dialogue Assistants for Long-term User-Agent Interaction (2025)
- MUSIC: MUlti-Step Instruction Contrast for Multi-Turn Reward Models (2025)
- Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents (2026)
- Mem-Gallery: Benchmarking Multimodal Long-Term Conversational Memory for MLLM Agents (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper