LLMs are General Asynchronous Agents

• 类型：arxiv
• 标识：2609.35427
• 链接：https://arxiv.org/abs/2609.35427
• 主分类：agent
• 形态：method
• TLDR：Modern LLMs are increasingly capable as autonomous agents, but they follow sequential interaction cycles: read, think, reply or call tools, repeat. Many real-world use cases are not sequential: voice assistants, embodied agents, and monitoring systems receive new inputs while they think or perform another task. Modern LLMs address this with specialized architectures for voice interaction and video streams, VLAs for robot control, asynchronous tool calling for API usage, and others. In this work, we generalize from different asynchronous tasks to general asynchronous agents that can adapt to di
• 待LLM分类：否
• 标题中文：[标题中文] LLM 是通用异步 Agent
• TLDR中文：现代 LLM 作为自主 Agent 的能力日益增强，但仍遵循顺序式交互循环：读取、思考、回复或调用工具、再重复。许多真实场景并非顺序式——语音助手、具身 Agent 和监控系统会在思考或执行其他任务时接收到新输入。现代 LLM 通过语音交互与视频流的专用架构、用于机器人控制的 VLA、用于 API 调用的异步工具调用等方式应对这一挑战。本工作将不同异步任务泛化为能适应不同场景的通用异步 Agent。
• 来源文件：
• /inbox/tom/_candidates/2026-09-30-agent-rag-longcontext-candidates.json