Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chai, Ziwei, Wang, Guoyin, Su, Jing, Zhang, Tianjie, Huang, Xuanwen, Wang, Xuwu, Xu, Jingjing, Yuan, Jianbo, Yang, Hongxia, Wu, Fei, Yang, Yang
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2403.16854
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909221501534208
author Chai, Ziwei
Wang, Guoyin
Su, Jing
Zhang, Tianjie
Huang, Xuanwen
Wang, Xuwu
Xu, Jingjing
Yuan, Jianbo
Yang, Hongxia
Wu, Fei
Yang, Yang
author_facet Chai, Ziwei
Wang, Guoyin
Su, Jing
Zhang, Tianjie
Huang, Xuanwen
Wang, Xuwu
Xu, Jingjing
Yuan, Jianbo
Yang, Hongxia
Wu, Fei
Yang, Yang
contents We present Expert-Token-Routing, a unified generalist framework that facilitates seamless integration of multiple expert LLMs. Our framework represents expert LLMs as special expert tokens within the vocabulary of a meta LLM. The meta LLM can route to an expert LLM like generating new tokens. Expert-Token-Routing not only supports learning the implicit expertise of expert LLMs from existing instruction dataset but also allows for dynamic extension of new expert LLMs in a plug-and-play manner. It also conceals the detailed collaboration process from the user's perspective, facilitating interaction as though it were a singular LLM. Our framework outperforms various existing multi-LLM collaboration paradigms across benchmarks that incorporate six diverse expert domains, demonstrating effectiveness and robustness in building generalist LLM system via synergizing multiple expert LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2403_16854
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle An Expert is Worth One Token: Synergizing Multiple Expert LLMs as Generalist via Expert Token Routing
Chai, Ziwei
Wang, Guoyin
Su, Jing
Zhang, Tianjie
Huang, Xuanwen
Wang, Xuwu
Xu, Jingjing
Yuan, Jianbo
Yang, Hongxia
Wu, Fei
Yang, Yang
Computation and Language
Artificial Intelligence
We present Expert-Token-Routing, a unified generalist framework that facilitates seamless integration of multiple expert LLMs. Our framework represents expert LLMs as special expert tokens within the vocabulary of a meta LLM. The meta LLM can route to an expert LLM like generating new tokens. Expert-Token-Routing not only supports learning the implicit expertise of expert LLMs from existing instruction dataset but also allows for dynamic extension of new expert LLMs in a plug-and-play manner. It also conceals the detailed collaboration process from the user's perspective, facilitating interaction as though it were a singular LLM. Our framework outperforms various existing multi-LLM collaboration paradigms across benchmarks that incorporate six diverse expert domains, demonstrating effectiveness and robustness in building generalist LLM system via synergizing multiple expert LLMs.
title An Expert is Worth One Token: Synergizing Multiple Expert LLMs as Generalist via Expert Token Routing
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2403.16854