ScrapMem: A Bio-inspired Framework for On-device Personalized Agent Memory via Optical Forgetting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910253217480704 |
|---|---|
| author | Chang, Jiale Ren, Yuxiang |
| author_facet | Chang, Jiale Ren, Yuxiang |
| contents | Long-term personalized memory for LLM agents is challenging on resource-limited edge devices due to high storage costs and multimodal complexity. To address this, we propose ScrapMem, a framework that integrates multimodal data into "Scrapbook Page." ScrapMem introduces Optical Forgetting, an optical compression mechanism that progressively reduces the resolution of older memories, lowering storage cost while suppressing low-value details. To maintain semantic consistency, we construct an Episodic Memory Graph (EM-Graph) that organizes key events into a causal-temporal structure. Extensive experiments on the multimodal ATM-Bench showcase that ScrapMem provides three main benefits: (1) strong performance, achieving a new state-of-the-art with a 51.0% Joint@10 score; (2) high storage efficiency, reducing memory usage by up to 93% via optical forgetting; and (3) improved recall, increasing Recall@10 to 70.3% through structured aggregation. ScrapMem offers an effective and storage-efficient solution for on-device long-term memory in multimodal LLM agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_03804 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ScrapMem: A Bio-inspired Framework for On-device Personalized Agent Memory via Optical Forgetting Chang, Jiale Ren, Yuxiang Artificial Intelligence I.2.7; I.2.11 Long-term personalized memory for LLM agents is challenging on resource-limited edge devices due to high storage costs and multimodal complexity. To address this, we propose ScrapMem, a framework that integrates multimodal data into "Scrapbook Page." ScrapMem introduces Optical Forgetting, an optical compression mechanism that progressively reduces the resolution of older memories, lowering storage cost while suppressing low-value details. To maintain semantic consistency, we construct an Episodic Memory Graph (EM-Graph) that organizes key events into a causal-temporal structure. Extensive experiments on the multimodal ATM-Bench showcase that ScrapMem provides three main benefits: (1) strong performance, achieving a new state-of-the-art with a 51.0% Joint@10 score; (2) high storage efficiency, reducing memory usage by up to 93% via optical forgetting; and (3) improved recall, increasing Recall@10 to 70.3% through structured aggregation. ScrapMem offers an effective and storage-efficient solution for on-device long-term memory in multimodal LLM agents. |
| title | ScrapMem: A Bio-inspired Framework for On-device Personalized Agent Memory via Optical Forgetting |
| topic | Artificial Intelligence I.2.7; I.2.11 |
| url | https://arxiv.org/abs/2605.03804 |