HyLaT: Efficient Multi-Agent Communication via Hybrid Latent-Text Protocol
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914599158153216 |
|---|---|
| author | Mou, Xinyi Wang, Siyuan Li, Zejun He, Yulan Wei, Zhongyu |
| author_facet | Mou, Xinyi Wang, Siyuan Li, Zejun He, Yulan Wei, Zhongyu |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_25421 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HyLaT: Efficient Multi-Agent Communication via Hybrid Latent-Text Protocol Mou, Xinyi Wang, Siyuan Li, Zejun He, Yulan Wei, Zhongyu Computation and Language 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. |
| title | HyLaT: Efficient Multi-Agent Communication via Hybrid Latent-Text Protocol |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2605.25421 |