arXiv cs.AI· Jiajun Zhu, Haobo Li, Zixin Chen, Huamin Qu, Wenshuo Zhang·· 12 小时前AI 评分20
LatBridge：潜空间通信多智能体系统中的高效稳定模型替换
LatBridge: Efficient and Stable Model Replacement in Latent-Communication Multi-Agent Systems
AI 导读
LatBridge 通过共享 Bridge 与独立接口组成的潜空间通信协议，实现多智能体系统在替换智能体模型后复用未变接口、匹配低秩接口并仅训练缺失部分，所有模型保持冻结。在八种小模型配置下，基础模式在 MATH 上平均准确率为 14.8%，在 ALFWorld 下一动作预测上为 27.3%。加速模式无需自回归生成即可构建目标输入。
正文 · 原文
Abstract:Multi-agent systems rely on communication to complete tasks, and latent communication passes continuous representations to reduce text generation and re-encoding. However, multi-agent systems must reconfigure agent models, whereas existing latent-communication methods are built for fixed model combinations, so their interfaces may fail after replacement. Our pilot study confirms this interface failure and identifies the requirement for interface reuse after model replacement. We present LatBridge, which connects agent models through a shared Bridge and separate interfaces under a shared latent-communication protocol. After agent model replacement, LatBridge reuses unchanged interfaces, selects matching low-rank interfaces, and combines them through the Bridge, training only missing ones with all models frozen. The Bridge processes source representations in the basic mode and constructs target input without autoregressive generation in the accelerated mode. Experiments support effective communication after model replacement, with mean basic-mode accuracies of 14.8% on MATH and 27.3% on ALFWorld next-action prediction across eight small-model configurations. Analyses examine the factors supporting task accuracy and communication efficiency.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.27813 [cs.AI] |
| (or arXiv:2610.27813v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.27813 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jiajun Zhu [view email]
[v1]
Tue, 20 Oct 2026 08:06:13 UTC (725 KB)
来源：arXiv cs.AI · arxiv.org