arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.01633cs.LGcs.AI
收藏

扩展图思维链推理:一种具有高效LLM服务机制的多智能体框架

Scaling Graph Chain-of-Thought Reasoning: A Multi-Agent Framework with Efficient LLM Serving

Nanjing University(南京大学);Ant Group(蚂蚁集团);University of New South Wales(新南威尔士大学);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Tsinghua University(清华大学)

Chengying Huan, Ziheng Meng, Yongchao Liu, Zhengyi Yang, Yun Zhu, Yue Yun, Shipeng Li, Rong Gu, Xiabao Wu, Haitao Zhang, Chuntao Hong, Shaonan Ma, Guihai Chen, Chen Tian

AI总结:

本文提出首个与优化LLM服务架构协同设计的多智能体Graph-CoT系统GLM,通过专用智能体分解推理并引入图特定KV缓存与流水线执行机制,大幅提升了推理准确率、降低了令牌成本与延迟,并显著提高了吞吐量。

摘要:

图思维链(Graph-CoT)使大型语言模型(LLM)能够对图结构知识进行逐步推理,但由于单智能体整体式提示、重复的上下文重编码以及低效的服务执行,现有流程存在准确率低、令牌消耗过多、延迟高和吞吐量低的问题。我们提出了GLM,这是首个与优化的LLM服务架构协同设计的多智能体Graph-CoT系统。GLM将推理分解为用于分类、推理、动作生成和图检索的专用智能体,实现分支和选择性上下文共享,以在保持推理质量的同时减少提示长度和推理迭代次数,从而提高准确率并降低总体令牌消耗。为了扩展推理,我们引入了一种感知Graph-CoT的LLM推理机制,结合图特定的KV缓存管理、基于优先级的驱逐和流水线执行,以提高服务效率。实验表明,与最先进的Graph-CoT基线相比,GLM将答案准确率提高了高达38%,令牌成本降低了高达95.7%,推理延迟降低了90.3%,并实现了高达15.1倍的吞吐量提升,从而能够大规模高效地应用于复杂的现实世界推理。

英文摘要:

Graph Chain-of-Thought (Graph-CoT) enables large language models (LLMs) to perform step-by-step reasoning over graph-structured knowledge, but existing pipelines suffer from low accuracy, excessive token usage, high latency, and low throughput due to single-agent monolithic prompts, repeated context re-encoding, and inefficient serving execution. We present GLM, the first multi-agent Graph-CoT system co-designed with an optimized LLM serving architecture. GLM decomposes reasoning into specialized agents for classification, reasoning, action generation, and graph retrieval, enabling branching and selective context sharing to reduce prompt length and reasoning iterations while preserving reasoning quality, thereby improving accuracy and reducing overall token consumption. To scale inference, we introduce a Graph-CoT-aware LLM inference mechanism with graph-specific KV-cache management, priority-based eviction, and pipelined execution to improve serving efficiency. Experiments demonstrate that GLM improves answer accuracy by up to 38%, reduces token cost by up to 95.7%, lowers inference latency by 90.3%, and achieves up to 15.1x higher throughput compared to state-of-the-art Graph-CoT baselines, enabling efficient adoption for complex real-world reasoning at scale.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑