Rethinking Memory in LLM based Agents: Representations, Operations, and Emerging Topics
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915693113376768 |
|---|---|
| author | Du, Yiming Huang, Wenyu Zheng, Danna Wang, Zhaowei Montella, Sebastien Lapata, Mirella Wong, Kam-Fai Pan, Jeff Z. |
| author_facet | Du, Yiming Huang, Wenyu Zheng, Danna Wang, Zhaowei Montella, Sebastien Lapata, Mirella Wong, Kam-Fai Pan, Jeff Z. |
| contents | Memory is fundamental to large language model (LLM)-based agents, but existing surveys emphasize application-level use (e.g., personalized dialogue), while overlooking the atomic operations governing memory dynamics. This work categorizes memory into parametric (implicit in model weights) and contextual (explicit external data, structured/unstructured) forms, and defines six core operations: Consolidation, Updating, Indexing, Forgetting, Retrieval, and Condensation. Mapping these dimensions reveals four key research topics: long-term, long-context, parametric modification, and multi-source memory. The taxonomy provides a structured view of memory-related research, benchmarks, and tools, clarifying functional interactions in LLM-based agents and guiding future advancements. The datasets, papers, and tools are publicly available at https://github.com/Elvin-Yiming-Du/Survey_Memory_in_AI. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_00675 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Rethinking Memory in LLM based Agents: Representations, Operations, and Emerging Topics Du, Yiming Huang, Wenyu Zheng, Danna Wang, Zhaowei Montella, Sebastien Lapata, Mirella Wong, Kam-Fai Pan, Jeff Z. Computation and Language Memory is fundamental to large language model (LLM)-based agents, but existing surveys emphasize application-level use (e.g., personalized dialogue), while overlooking the atomic operations governing memory dynamics. This work categorizes memory into parametric (implicit in model weights) and contextual (explicit external data, structured/unstructured) forms, and defines six core operations: Consolidation, Updating, Indexing, Forgetting, Retrieval, and Condensation. Mapping these dimensions reveals four key research topics: long-term, long-context, parametric modification, and multi-source memory. The taxonomy provides a structured view of memory-related research, benchmarks, and tools, clarifying functional interactions in LLM-based agents and guiding future advancements. The datasets, papers, and tools are publicly available at https://github.com/Elvin-Yiming-Du/Survey_Memory_in_AI. |
| title | Rethinking Memory in LLM based Agents: Representations, Operations, and Emerging Topics |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2505.00675 |