dorsal/arxiv
View SchemaHiMem: Hierarchical Long-Term Memory for LLM Long-Horizon Agents
| Authors | Ningning Zhang, Xingxing Yang, Zhizhong Tan, Weiping Deng, Wenyong Wang |
|---|---|
| Categories | |
| ArXiv ID | 2601.06377vv1 |
| URL | https://arxiv.org/abs/2601.06377 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Although long-term memory systems have made substantial progress in recent years, they still exhibit clear limitations in adaptability, scalability, and self-evolution under continuous interaction settings. Inspired by cognitive theories, we propose HiMem, a hierarchical long-term memory framework for long-horizon dialogues, designed to support memory construction, retrieval, and dynamic updating during sustained interactions. HiMem constructs cognitively consistent Episode Memory via a Topic-Aware Event--Surprise Dual-Channel Segmentation strategy, and builds Note Memory that captures stable knowledge through a multi-stage information extraction pipeline. These two memory types are semantically linked to form a hierarchical structure that bridges concrete interaction events and abstract knowledge, enabling efficient retrieval without sacrificing information fidelity. HiMem supports both hybrid and best-effort retrieval strategies to balance accuracy and efficiency, and incorporates conflict-aware Memory Reconsolidation to revise and supplement stored knowledge based on retrieval feedback. This design enables continual memory self-evolution over long-term use. Experimental results on long-horizon dialogue benchmarks demonstrate that HiMem consistently outperforms representative baselines in accuracy, consistency, and long-term reasoning, while maintaining favorable efficiency. Overall, HiMem provides a principled and scalable design paradigm for building adaptive and self-evolving LLM-based conversational agents. The code is available at https://github.com/jojopdq/HiMem.
{
"annotation_id": "d50dbe2e-3a14-4b20-8d85-58ed572db13f",
"date_created": "2026-02-17T05:53:07.564000Z",
"date_modified": "2026-02-17T05:53:07.564000Z",
"file_hash": "26c08a417bf5ec96663a508dd54b3d8b8dfc5673ba05b7895dc219ee1e7761c7",
"private": false,
"record": {
"abstract": "Although long-term memory systems have made substantial progress in recent years, they still exhibit clear limitations in adaptability, scalability, and self-evolution under continuous interaction settings. Inspired by cognitive theories, we propose HiMem, a hierarchical long-term memory framework for long-horizon dialogues, designed to support memory construction, retrieval, and dynamic updating during sustained interactions. HiMem constructs cognitively consistent Episode Memory via a Topic-Aware Event--Surprise Dual-Channel Segmentation strategy, and builds Note Memory that captures stable knowledge through a multi-stage information extraction pipeline. These two memory types are semantically linked to form a hierarchical structure that bridges concrete interaction events and abstract knowledge, enabling efficient retrieval without sacrificing information fidelity. HiMem supports both hybrid and best-effort retrieval strategies to balance accuracy and efficiency, and incorporates conflict-aware Memory Reconsolidation to revise and supplement stored knowledge based on retrieval feedback. This design enables continual memory self-evolution over long-term use. Experimental results on long-horizon dialogue benchmarks demonstrate that HiMem consistently outperforms representative baselines in accuracy, consistency, and long-term reasoning, while maintaining favorable efficiency. Overall, HiMem provides a principled and scalable design paradigm for building adaptive and self-evolving LLM-based conversational agents. The code is available at https://github.com/jojopdq/HiMem.",
"arxiv_id": "2601.06377",
"authors": [
"Ningning Zhang",
"Xingxing Yang",
"Zhizhong Tan",
"Weiping Deng",
"Wenyong Wang"
],
"categories": [
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "HiMem: Hierarchical Long-Term Memory for LLM Long-Horizon Agents",
"url": "https://arxiv.org/abs/2601.06377",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "ae095e0e-3de9-411a-b029-dd13066c0dea",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}