AgentDojo: 用于评估LLM代理攻击与防御的动态环境
论文AgentDojo: A Dynamic Environment to Evaluate Attacks […]
AgentDojo: 用于评估LLM代理攻击与防御的动态环境 Read More »
论文AgentDojo: A Dynamic Environment to Evaluate Attacks […]
AgentDojo: 用于评估LLM代理攻击与防御的动态环境 Read More »
论文SciAgents: Automating scientific discovery through mu
SciAgents:通过多代理系统(Multi-Agent System)和知识图谱(Knowledge Graphs),实现科学发现的自动化 Read More »
论文On the Limits of Agency in Agent-Based Models(《论基于代理的
AgentTorch:扩展基于代理的模型的新型框架,通过高效的计算方式可实现对百万级别代理的模拟 Read More »
近期,Google DeepMind的联合创始人兼CEO Demis Hassabis与主持人Hannah F
Unreasonably Effective AI ,“超乎寻常有效”或“不合常理地有效”的AI Read More »
论文Diffusion Models Are Real-Time Game Engines(《扩散模型是实时游
GameNGen:首个完全由神经网络模型驱动的游戏引擎 Read More »
论文《A Review of Large Language Models and Autonomous Age
化学中的LLMs,Large Language Models and Autonomous Agents in Chemistry Read More »