Jones Ray

ScholarPulse 日报 2026-09-16

2026-09-16 学术简报：2 篇。智能体社会中，即使诚实且有能力的智能体在现有通信机制下也常无法达成满意结果，需要一种超越个体"个人约束"的"社会约束框架"来保障协作安全。

今日速览

序号标题来源日期主题推荐等级
1Agentic Societies Need a Social HarnessarXiv2026-09-15AI-Agent高
2FlashVector: Agent for Hierarchical Model Serving Stack OptimizationarXiv2026-09-15AI-Agent高

重点论文与技术动态

1. Agentic Societies Need a Social Harness

一句话结论

智能体社会中，即使诚实且有能力的智能体在现有通信机制下也常无法达成满意结果，需要一种超越个体”个人约束”的”社会约束框架”来保障协作安全。

核心内容

方法与数据

价值判断

摘要 An agentic society is a collection of AI agents that coordinate autonomously across trust boundaries, on behalf of different principals whose objectives may only partially align. We show experimentally that in agentic societies even honest, competent agents often fail to reach satisfactory outcomes with existing harnesses and messaging primitives, and that faulty or malicious agents can stall collaboration, influence outcomes, and pursue other harmful goals by exploiting vulnerabilities in communication (``speech''). We argue that agentic societies need a \emph{social harness} for inter-agent interactions, in addition to each agent's \emph{personal harness}, which manages its private context and communication with its principal. We propose a layered architecture for social harnesses which (i) prevents classes of failures outright, (ii) enables agents to detect invalid messages at runtime, and (iii) supports post-facto investigation and consequences, and highlight directions for future research to realize these capabilities.

2. FlashVector: Agent for Hierarchical Model Serving Stack Optimization

一句话结论

FlashVector 是一个跨层模型服务栈优化 Agent 系统，在 Unity 广告平台部署后实现了最高 2 倍吞吐量和 1.98 倍延迟加速。

核心内容

方法与数据

价值判断

摘要 Model serving is one of the largest cost drivers in production recommender systems. Maximizing its throughput requires navigating a deeply layered hierarchy: GPU kernels, the ML framework computation graph, the model server, and on-demand feature processing -- each demanding specialized domain expertise. Such cross-layer expertise is inherently difficult to acquire, and does not scale with a workload that continuously grows and evolves, leaving significant cost efficiency gains unrealized. While recent AI agents have demonstrated human expert level efficiency in standalone GPU kernel optimization, automated tuning and optimization for the rest of the serving stack remain largely unexplored. We present FlashVector, an agentic system that optimizes performance across all layers of the model serving stack. The key contribution is an extensible framework to generalize the single kernel optimization agent paradigm to heterogeneous technical stacks, and to deliver performance improvements holistically. After deployment in Unity's Vector advertising platform, FlashVector achieved up to 2x throughput increase and up to 1.98x latency speedup on model server, and up to 1.6x throughput increase on feature store. These optimizations were discovered not only at the GPU kernel and computation graph levels, but also across the other components of the model serving stack, such as the model server (NVIDIA Triton's C++ codebase) and the on-demand feature transformation service (Python codebase), demonstrating the extensibility of the framework to more complex system architectures.