A Structure-Agnostic Co-Tuning Framework for LLMs and SLMs in Cloud-Edge Systems
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914158917713920 |
|---|---|
| author | Liu, Yuze Wang, Yunhan Zhang, Tiehua Shen, Zhishu Peng, Cheng Wu, Libing Xia, Feng Jin, Jiong |
| author_facet | Liu, Yuze Wang, Yunhan Zhang, Tiehua Shen, Zhishu Peng, Cheng Wu, Libing Xia, Feng Jin, Jiong |
| contents | The surge in intelligent applications driven by large language models (LLMs) has made it increasingly difficult for bandwidth-limited cloud servers to process extensive LLM workloads in real time without compromising user data privacy. To solve these problems, recent research has focused on constructing cloud-edge consortia that integrate server-based LLM with small language models (SLMs) on mobile edge devices. Furthermore, designing collaborative training mechanisms within such consortia to enhance inference performance has emerged as a promising research direction. However, the cross-domain deployment of SLMs, coupled with structural heterogeneity in SLMs architectures, poses significant challenges to enhancing model performance. To this end, we propose Co-PLMs, a novel co-tuning framework for collaborative training of large and small language models, which integrates the process of structure-agnostic mutual learning to realize knowledge exchange between the heterogeneous language models. This framework employs distilled proxy models (DPMs) as bridges to enable collaborative training between the heterogeneous server-based LLM and on-device SLMs, while preserving the domain-specific insights of each device. The experimental results show that Co-PLMs outperform state-of-the-art methods, achieving average increases of 5.38% in Rouge-L and 4.88% in EM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_11678 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Structure-Agnostic Co-Tuning Framework for LLMs and SLMs in Cloud-Edge Systems Liu, Yuze Wang, Yunhan Zhang, Tiehua Shen, Zhishu Peng, Cheng Wu, Libing Xia, Feng Jin, Jiong Distributed, Parallel, and Cluster Computing Artificial Intelligence Computation and Language Machine Learning The surge in intelligent applications driven by large language models (LLMs) has made it increasingly difficult for bandwidth-limited cloud servers to process extensive LLM workloads in real time without compromising user data privacy. To solve these problems, recent research has focused on constructing cloud-edge consortia that integrate server-based LLM with small language models (SLMs) on mobile edge devices. Furthermore, designing collaborative training mechanisms within such consortia to enhance inference performance has emerged as a promising research direction. However, the cross-domain deployment of SLMs, coupled with structural heterogeneity in SLMs architectures, poses significant challenges to enhancing model performance. To this end, we propose Co-PLMs, a novel co-tuning framework for collaborative training of large and small language models, which integrates the process of structure-agnostic mutual learning to realize knowledge exchange between the heterogeneous language models. This framework employs distilled proxy models (DPMs) as bridges to enable collaborative training between the heterogeneous server-based LLM and on-device SLMs, while preserving the domain-specific insights of each device. The experimental results show that Co-PLMs outperform state-of-the-art methods, achieving average increases of 5.38% in Rouge-L and 4.88% in EM. |
| title | A Structure-Agnostic Co-Tuning Framework for LLMs and SLMs in Cloud-Edge Systems |
| topic | Distributed, Parallel, and Cluster Computing Artificial Intelligence Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2511.11678 |