arXiv cs.AI· Siqi Xiang, Zhipeng Xu, Shengrui Liu, Zulong Chen, Chunyan Miao, Shijian Lu·· 10 小时前AI 评分38
把分辨率当注意力:RAA 让 LLM 智能体记忆感知更省 token
Resolution as Attention for Efficient Agent Memory Perception
AI 导读
研究者提出 Resolution as Attention(RAA)记忆感知机制,将历史交互渲染为图像,并按与下一步决策的相关性动态调整分辨率,把视觉分辨率当作注意力分配手段。实验显示 RAA 在多个任务和骨干模型上将 token 消耗降低 50% 以上且不损害任务表现,还可作为插件感知层与现有记忆管理方法结合。
正文 · 原文
Abstract:LLM-based agents accumulate extensive interaction histories when solving complex, long-horizon tasks, making efficient memory management critical to their performance and scalability. While existing memory management methods primarily focus on what to retain or retrieve, they often overlook how much perceptual detail should be allocated to individual past interactions. We posit that effective agent memory depends not only on the information preserved, but also on the granularity at which each past interaction is perceived. Motivated by this insight, we propose Resolution as Attention (RAA), a memory perception mechanism that uses visual resolution as a means of attention allocation across interaction histories. RAA renders past interactions as images and dynamically adjusts their resolution at every step according to their relevance to the agent's upcoming decision, preserving fine-grained information for critical interactions while reducing visual-token consumption for less relevant ones. To adaptively determine the perceptual fidelity of each interaction, we train a lightweight relevance estimator with teacher-model guidance and further optimize it to preserve the agent's decision consistency at a reduced visual-token cost. Experiments across multiple tasks and backbones show that RAA reduces token consumption by over 50% without compromising task performance. Moreover, RAA can be integrated as a plug-in perception layer with existing memory management methods, yielding further token savings and revealing adaptive perception as a complementary dimension of agent memory.
| Subjects: | Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2611.24422 [cs.AI] |
| (or arXiv:2611.24422v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.24422 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Siqi Xiang [view email]
[v1]
Wed, 18 Nov 2026 13:43:16 UTC (950 KB)
来源:arXiv cs.AI · arxiv.org