2026年9月15日
What Deserves Memory: Adaptive Memory Distillation for LLM Agents
摘要： 论文阅读：What Deserves Memory——什么信息真正值得 Agent 记住？ 论文标题：What Deserves Memory: Adaptive Memory Distillation for LLM Agents 作者：Wenquan Ma, Jiayan Nan, Wenlon 阅读全文
posted @ 2026-09-15 10:28 YourF4u1t 阅读(18) 评论(0) 推荐(0)
GAM: Hierarchical Graph-based Agentic Memory for LLM Agents
摘要： 论文阅读：GAM——面向 LLM Agent 的层次图式记忆 论文标题：GAM: Hierarchical Graph-based Agentic Memory for LLM Agents 作者：Zhaofen Wu, Hanrong Zhang, Fulin Lin, Wujiang Xu, X 阅读全文
posted @ 2026-09-15 10:18 YourF4u1t 阅读(17) 评论(0) 推荐(0)
Chain-of-Memory: Lightweight Memory Construction with Dynamic Evolution for LLM Agents
摘要： 论文阅读：Chain-of-Memory——用动态记忆链替代复杂的 Memory Construction 论文标题：Chain-of-Memory: Lightweight Memory Construction with Dynamic Evolution for LLM Agents 作者：X 阅读全文
posted @ 2026-09-15 10:11 YourF4u1t 阅读(12) 评论(0) 推荐(0)
MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents
摘要： 论文阅读：MAGMA——面向 AI Agent 的多图 Agentic Memory 架构 论文标题：MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents 作者：Dongming Jiang, Yi Li, Guan 阅读全文
posted @ 2026-09-15 10:06 YourF4u1t 阅读(13) 评论(0) 推荐(0)
Does Memory Need Graphs? A Unified Framework and Empirical Analysis for Long-Term Dialog Memory
摘要： 论文阅读：Does Memory Need Graphs? —— 长期对话 Memory 真的需要图结构吗？ 论文标题：Does Memory Need Graphs? A Unified Framework and Empirical Analysis for Long-Term Dialog M 阅读全文
posted @ 2026-09-15 09:47 YourF4u1t 阅读(9) 评论(0) 推荐(0)
2026年9月14日
How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior
摘要： 论文阅读：Memory 管理如何影响 LLM Agent——经验跟随行为的实证研究 论文标题：How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior 作者：Zidi X 阅读全文
posted @ 2026-09-14 20:41 YourF4u1t 阅读(19) 评论(0) 推荐(0)
StructMem: Structured Memory for Long-Horizon Behavior in LLMs
摘要： 论文阅读：StructMem——面向 LLM 长程行为的结构化记忆 论文标题：StructMem: Structured Memory for Long-Horizon Behavior in LLMs 作者：Buqiang Xu、Yijun Chen、Jizhan Fang、Ruobin Zhon 阅读全文
posted @ 2026-09-14 14:38 YourF4u1t 阅读(13) 评论(0) 推荐(0)
Lightweight LLM Agent Memory with Small Language Models
摘要： 论文阅读：LightMem——使用小语言模型构建轻量级 LLM Agent Memory 论文标题：Lightweight LLM Agent Memory with Small Language Models 作者：Jiaquan Zhang, Chaoning Zhang, Shuxu Chen 阅读全文
posted @ 2026-09-14 13:21 YourF4u1t 阅读(13) 评论(0) 推荐(0)
2026年9月13日
Agent Data Injection Attacks are Realistic Threats to AI Agents
摘要： 论文阅读：Agent 数据注入攻击是 AI Agent 的现实威胁 论文标题：Agent Data Injection Attacks are Realistic Threats to AI Agents 作者：Woohyuk Choi、Juhee Kim、Taehyun Kang、Jihyeon 阅读全文
posted @ 2026-09-13 22:50 YourF4u1t 阅读(19) 评论(0) 推荐(0)
Self-Attribution Bias: When AI Monitors Go Easy on Themselves
摘要： 论文阅读：Self-Attribution Bias——当 AI 监控器对自己手下留情 论文标题：Self-Attribution Bias: When AI Monitors Go Easy on Themselves 中文标题：自我归因偏误：当 AI 监控器对自己手下留情 作者：Dipika K 阅读全文
posted @ 2026-09-13 18:10 YourF4u1t 阅读(18) 评论(0) 推荐(0)