Enabling Flexible Multi-LLM Integration for Scalable Knowledge Aggregation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Kong, Zhenglun, Zhan, Zheng, Hou, Shiyue, Gong, Yifan, Meng, Xin, Sui, Pengwei, Dong, Peiyan, Shen, Xuan, Wang, Zifeng, Zhao, Pu, Tang, Hao, Ioannidis, Stratis, Wang, Yanzhi
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913866485596160
author Kong, Zhenglun
Zhan, Zheng
Hou, Shiyue
Gong, Yifan
Meng, Xin
Sui, Pengwei
Dong, Peiyan
Shen, Xuan
Wang, Zifeng
Zhao, Pu
Tang, Hao
Ioannidis, Stratis
Wang, Yanzhi
author_facet Kong, Zhenglun
Zhan, Zheng
Hou, Shiyue
Gong, Yifan
Meng, Xin
Sui, Pengwei
Dong, Peiyan
Shen, Xuan
Wang, Zifeng
Zhao, Pu
Tang, Hao
Ioannidis, Stratis
Wang, Yanzhi
contents Large language models (LLMs) have shown remarkable promise but remain challenging to continually improve through traditional finetuning, particularly when integrating capabilities from other specialized LLMs. Popular methods like ensemble and weight merging require substantial memory and struggle to adapt to changing data environments. Recent efforts have transferred knowledge from multiple LLMs into a single target model; however, they suffer from interference and degraded performance among tasks, largely due to limited flexibility in candidate selection and training pipelines. To address these issues, we propose a framework that adaptively selects and aggregates knowledge from diverse LLMs to build a single, stronger model, avoiding the high memory overhead of ensemble and inflexible weight merging. Specifically, we design an adaptive selection network that identifies the most relevant source LLMs based on their scores, thereby reducing knowledge interference. We further propose a dynamic weighted fusion strategy that accounts for the inherent strengths of candidate LLMs, along with a feedback-driven loss function that prevents the selector from converging on a single subset of sources. Experimental results demonstrate that our method can enable a more stable and scalable knowledge aggregation process while reducing knowledge interference by up to 50% compared to existing approaches. Code is avaliable at https://github.com/ZLKong/LLM_Integration
format Preprint
id arxiv_https___arxiv_org_abs_2505_23844
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enabling Flexible Multi-LLM Integration for Scalable Knowledge Aggregation
Kong, Zhenglun
Zhan, Zheng
Hou, Shiyue
Gong, Yifan
Meng, Xin
Sui, Pengwei
Dong, Peiyan
Shen, Xuan
Wang, Zifeng
Zhao, Pu
Tang, Hao
Ioannidis, Stratis
Wang, Yanzhi
Computation and Language
Large language models (LLMs) have shown remarkable promise but remain challenging to continually improve through traditional finetuning, particularly when integrating capabilities from other specialized LLMs. Popular methods like ensemble and weight merging require substantial memory and struggle to adapt to changing data environments. Recent efforts have transferred knowledge from multiple LLMs into a single target model; however, they suffer from interference and degraded performance among tasks, largely due to limited flexibility in candidate selection and training pipelines. To address these issues, we propose a framework that adaptively selects and aggregates knowledge from diverse LLMs to build a single, stronger model, avoiding the high memory overhead of ensemble and inflexible weight merging. Specifically, we design an adaptive selection network that identifies the most relevant source LLMs based on their scores, thereby reducing knowledge interference. We further propose a dynamic weighted fusion strategy that accounts for the inherent strengths of candidate LLMs, along with a feedback-driven loss function that prevents the selector from converging on a single subset of sources. Experimental results demonstrate that our method can enable a more stable and scalable knowledge aggregation process while reducing knowledge interference by up to 50% compared to existing approaches. Code is avaliable at https://github.com/ZLKong/LLM_Integration
title Enabling Flexible Multi-LLM Integration for Scalable Knowledge Aggregation
topic Computation and Language
url https://arxiv.org/abs/2505.23844