返回全部动态
多智能体LLM系统中推理时并行的两层视角
原标题：A Two-Tier Perspective on Inference-Time Parallelism in Multi-Agent LLM Systems
AI 摘要
该论文提出了一种名为TIPEX的可控执行框架，用于统一多智能体LLM系统中的副本并行和结构并行两种推理时并行级别。实验表明，推理时并行能显著提高准确率并降低端到端延迟，但会增加令牌消耗。进一步分析发现，副本并行和结构并行在不同任务复杂度下具有互补效应，中等难度任务受益最大，而过度的并行策略并不一定带来更好的性能。
以上摘要由 AI 生成，可能存在误差。事实请以原文为准。
正文节选
Computer Science > Multiagent Systems Title:A Two-Tier Perspective on Inference-Time Parallelism in Multi-Agent LLM Systems View PDF HTML (experimental) Abstract:Large language model (LLM)-driven multi-agent systems typically require multiple model invocations and complex coordination during inference, and their execution strategies directly affect system accuracy, latency, and computational cost. Parallel execution provides a means to improve inference-time efficiency. From the pers
发布时间：2026-08-07 12:00
抓取时间：2026-08-07 14:32
来源机构：arXiv