MEGA Hub

First Demonstration of Multi-Agent LLM System for Million-Scale Optical Link Management in Global Production AIDCs

Authors

Do you know Jingyi Su?You can claim authorship or link another user.Do you know Yihao Zhang?You can claim authorship or link another user.Do you know Dianxuan Fu?You can claim authorship or link another user.Do you know Leiyan Fei?You can claim authorship or link another user.Do you know Juan Wang?You can claim authorship or link another user.Do you know Mengfan Dai?You can claim authorship or link another user.Do you know Qing Liu?You can claim authorship or link another user.Do you know Xiong Wu?You can claim authorship or link another user.Do you know Yufeng Jiang?You can claim authorship or link another user.Do you know Cheng Chen?You can claim authorship or link another user.Do you know Bowen Zhang?You can claim authorship or link another user.Do you know Peilong Wang?You can claim authorship or link another user.Do you know Xi Chen?You can claim authorship or link another user.Do you know Zonglong He?You can claim authorship or link another user.Do you know Hongchen Yu?You can claim authorship or link another user.Do you know Zhicheng Ye?You can claim authorship or link another user.Do you know Weisheng Hu?You can claim authorship or link another user.Do you know Qunbi Zhuge?You can claim authorship or link another user.

Abstract

We present the first LLM-powered multi-agent system for autonomous fault management across millions of optical links in production AIDCs. Refined via SFT and continuous memory evolution, it achieves 97.7% F1 and over 60% fault-incident reduction, outperforming SOTA LLMs on a ten-week field data evaluation.

Community

00

Publication notes

Author note
4 pages, 3 figures