Choosing How to Remember: Adaptive Memory Structures for LLM Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915799548035072 |
|---|---|
| author | Lu, Mingfei Wu, Mengjia Liu, Feng Xu, Jiawei Li, Weikai Wang, Haoyang Hu, Zhengdong Ding, Ying Sun, Yizhou Lu, Jie Zhang, Yi |
| author_facet | Lu, Mingfei Wu, Mengjia Liu, Feng Xu, Jiawei Li, Weikai Wang, Haoyang Hu, Zhengdong Ding, Ying Sun, Yizhou Lu, Jie Zhang, Yi |
| contents | Memory is critical for enabling large language model (LLM) based agents to maintain coherent behavior over long-horizon interactions. However, existing agent memory systems suffer from two key gaps: they rely on a one-size-fits-all memory structure and do not model memory structure selection as a context-adaptive decision, limiting their ability to handle heterogeneous interaction patterns and resulting in suboptimal performance. We propose a unified framework, FluxMem, that enables adaptive memory organization for LLM agents. Our framework equips agents with multiple complementary memory structures. It explicitly learns to select among these structures based on interaction-level features, using offline supervision derived from downstream response quality and memory utilization. To support robust long-horizon memory evolution, we further introduce a three-level memory hierarchy and a Beta Mixture Model-based probabilistic gate for distribution-aware memory fusion, replacing brittle similarity thresholds. Experiments on two long-horizon benchmarks, PERSONAMEM and LoCoMo, demonstrate that our method achieves average improvements of 9.18% and 6.14%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_14038 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Choosing How to Remember: Adaptive Memory Structures for LLM Agents Lu, Mingfei Wu, Mengjia Liu, Feng Xu, Jiawei Li, Weikai Wang, Haoyang Hu, Zhengdong Ding, Ying Sun, Yizhou Lu, Jie Zhang, Yi Artificial Intelligence Machine Learning Memory is critical for enabling large language model (LLM) based agents to maintain coherent behavior over long-horizon interactions. However, existing agent memory systems suffer from two key gaps: they rely on a one-size-fits-all memory structure and do not model memory structure selection as a context-adaptive decision, limiting their ability to handle heterogeneous interaction patterns and resulting in suboptimal performance. We propose a unified framework, FluxMem, that enables adaptive memory organization for LLM agents. Our framework equips agents with multiple complementary memory structures. It explicitly learns to select among these structures based on interaction-level features, using offline supervision derived from downstream response quality and memory utilization. To support robust long-horizon memory evolution, we further introduce a three-level memory hierarchy and a Beta Mixture Model-based probabilistic gate for distribution-aware memory fusion, replacing brittle similarity thresholds. Experiments on two long-horizon benchmarks, PERSONAMEM and LoCoMo, demonstrate that our method achieves average improvements of 9.18% and 6.14%. |
| title | Choosing How to Remember: Adaptive Memory Structures for LLM Agents |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2602.14038 |