arXiv cs.CL· Xinyi Mou, Siyuan Wang, Zejun Li, Yulan He, Zhongyu Wei·· 14 小时前AI 评分31
HyLaT：用混合隐空间-文本协议实现高效多智能体通信
HyLaT: Efficient Multi-Agent Communication via Hybrid Latent-Text Protocol
AI 导读
HyLaT 是一种混合隐空间-文本通信协议，通过隐空间通道传输复杂认知信号以提升效率，同时用自然语言表达关键信号以保持可解释性与精度。该协议采用两阶段训练框架，结合单智能体混合生成学习与多智能体交互协同训练，使智能体能在多轮交互中生成并理解混合消息。实验显示 HyLaT 显著降低通信开销，同时保持有竞争力的任务表现，并在多种设置下具备较强泛化性与鲁棒性。
正文 · 原文
Abstract:Communication protocol design is a central challenge in large language model-based multi-agent systems. Existing single-channel approaches face an inherent communication trilemma: text-based methods are interpretable but verbose, while latent-space methods are efficient but opaque and limited to unidirectional workflows. Inspired by multi-channel communication theory, we propose HyLaT, a hybrid latent-text communication protocol that transmits elaborate cognitive signals through a latent channel for efficiency, while expressing concise critical signals in natural language to preserve interpretability and precision. We introduce a two-stage training framework combining single-agent hybrid generation learning and multi-agent interactive co-training, enabling agents to generate and interpret hybrid messages across multiple rounds of interaction. Experiments demonstrate that HyLaT reduces communication overhead significantly while maintaining competitive task performance, with strong generalization and robustness across diverse settings.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2605.25421 [cs.CL] |
| (or arXiv:2605.25421v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.25421 arXiv-issued DOI via DataCite |
Submission history
From: Xinyi Mou [view email]
[v1]
Mon, 25 May 2026 04:50:15 UTC (948 KB)
[v2]
Fri, 23 Oct 2026 19:02:23 UTC (482 KB)
来源：arXiv cs.CL · arxiv.org