MeMo: Memory as a Model
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_ | 1866911702074785792 |
|---|---|
| author | Quek, Ryan Wei Heng Lee, Sanghyuk Leong, Alfred Wei Lun Verma, Arun Prakash, Alok Chen, Nancy F. Low, Bryan Kian Hsiang Rus, Daniela Solar-Lezama, Armando |
| author_facet | Quek, Ryan Wei Heng Lee, Sanghyuk Leong, Alfred Wei Lun Verma, Arun Prakash, Alok Chen, Nancy F. Low, Bryan Kian Hsiang Rus, Daniela Solar-Lezama, Armando |
| contents | Large language models (LLMs) achieve strong performance across a wide range of tasks, but remain frozen after pretraining until subsequent updates. Many real-world applications require timely, domain-specific information, motivating the need for efficient mechanisms to incorporate new knowledge. In this paper, we introduce MeMo (Memory as a Model), a modular framework that encodes new knowledge into a dedicated memory model while keeping the LLM parameters unchanged. Compared to existing methods, MeMo offers several advantages: (a) it captures complex cross-document relationships, (b) it is robust to retrieval noise, (c) it avoids catastrophic forgetting in the LLM, (d) it does not require access to the LLM's weights or output logits, enabling plug-and-play integration with both open and proprietary closed-source LLMs, and (e) its retrieval cost is independent of corpus size at inference time. Our experimental results on three benchmarks, BrowseComp-Plus, NarrativeQA, and MuSiQue, show that MeMo achieves strong performance compared to existing methods across diverse settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_15156 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MeMo: Memory as a Model Quek, Ryan Wei Heng Lee, Sanghyuk Leong, Alfred Wei Lun Verma, Arun Prakash, Alok Chen, Nancy F. Low, Bryan Kian Hsiang Rus, Daniela Solar-Lezama, Armando Computation and Language Artificial Intelligence Machine Learning Large language models (LLMs) achieve strong performance across a wide range of tasks, but remain frozen after pretraining until subsequent updates. Many real-world applications require timely, domain-specific information, motivating the need for efficient mechanisms to incorporate new knowledge. In this paper, we introduce MeMo (Memory as a Model), a modular framework that encodes new knowledge into a dedicated memory model while keeping the LLM parameters unchanged. Compared to existing methods, MeMo offers several advantages: (a) it captures complex cross-document relationships, (b) it is robust to retrieval noise, (c) it avoids catastrophic forgetting in the LLM, (d) it does not require access to the LLM's weights or output logits, enabling plug-and-play integration with both open and proprietary closed-source LLMs, and (e) its retrieval cost is independent of corpus size at inference time. Our experimental results on three benchmarks, BrowseComp-Plus, NarrativeQA, and MuSiQue, show that MeMo achieves strong performance compared to existing methods across diverse settings. |
| title | MeMo: Memory as a Model |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2605.15156 |