深度探讨LLM代理在医疗场景中的可信赖性问题，为AI医疗安全提供关键研究视角。
arXiv:2607.05055v1 Announce Type: new Abstract: Healthcare appointment scheduling remains a persistent operational bottleneck, driven by manual coordi…
揭示嵌入防御在LLM多代理系统中的根本性失效，提出安全范式新思考
arXiv:2605.01133v2 Announce Type: replace-cross Abstract: Large language model (LLM)-powered multi-agent systems (MAS) enable agents to communicate an…
开源与专有AI代理评审系统对比评测，揭示AI辅助科研时代评审新挑战
arXiv:2606.19749v1 Announce Type: new Abstract: A new class of agentic review systems are emerging as a remedy to the pressure placed on peer review s…
AI Agent的每个功能都是潜在缓存失效点，揭示RAG与多代理架构的陷阱与反思
Article URL: https://www.openclacky.com/engineering/cache-invalidation-surface Comments URL: https://news.ycombinator.com/item?id=48426050 Points: 2 #…
工业级GenAI评估利器：GLIDE库如何实现可靠的预测驱动推理与代理系统评测。
arXiv:2605.31278v1 Announce Type: new Abstract: Reliable evaluation of agentic systems requires unbiased estimates with valid uncertainty, but standar…
下一代LLM代理能否超越前代的合作偏见？跨提供商的实证博弈揭示多智能体系统的均衡新面貌。
arXiv:2605.29874v1 Announce Type: cross Abstract: Do next-generation LLM agents inherit the cooperative biases documented in their predecessors, or do…
专为竞争保护打造的法律AI研究助手，能智能检索与分析法规，效率惊人！
arXiv:2605.27331v1 Announce Type: new Abstract: Competition law experts conducting legal research must review extensive volumes of cases, decisions, a…
架构升级比提示工程更有效：7节点专业路由多智能体系统在临床基准上实现飞跃
arXiv:2605.24699v1 Announce Type: new Abstract: Most reported gains on agentic-LLM clinical benchmarks are often attributed to prompt engineering, yet…
首次揭示LLM代理系统中技能缩放定律，基于15个模型、超百万数据点的实证研究
arXiv:2605.16508v1 Announce Type: new Abstract: As agent systems scale, skills accumulate into large reusable libraries, yet their scaling laws remain…