# A Structure-Agnostic Co-Tuning Framework for LLMs and SLMs in Cloud-Edge Systems

arXiv: https://arxiv.org/abs/2511.11678 · HTML: https://mlsanity.com/paper/2511.11678/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 48 references that are in the map.
- First posted on arXiv 12 Nov 2025.

## Builds on (most cited first, 17 of 17)

- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks](https://mlsanity.com/paper/2204.07705/) (2022)
- [TinyLlama: An Open-Source Small Language Model](https://mlsanity.com/paper/2401.02385/) (2024)
- [Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning](https://mlsanity.com/paper/2310.06694/) (2023)
- [SecureBoost: A Lossless Federated Learning Framework](https://mlsanity.com/paper/1901.08755/) (2019)
- [Qwen2.5-1M Technical Report](https://mlsanity.com/paper/2501.15383/) (2025)
- [FedPrompt: Communication-Efficient and Privacy Preserving Prompt Tuning in Federated Learning](https://mlsanity.com/paper/2208.12268/) (2022)
- [Tag-LLM: Repurposing General-Purpose LLMs for Specialized Domains](https://mlsanity.com/paper/2402.05140/) (2024)
- [CE-CoLLM: Efficient and Adaptive Large Language Models Through Cloud-Edge Collaboration](https://mlsanity.com/paper/2411.02829/) (2024)
- [FedPFT: Federated Proxy Fine-Tuning of Foundation Models](https://mlsanity.com/paper/2404.11536/) (2024)
- [Towards Heterogeneity-Aware and Energy-Efficient Topology Optimization for Decentralized Federated Learning in Edge Environment](https://mlsanity.com/paper/2508.08278/) (2025)
- [GPT4Image: Large Pre-trained Models Help Vision Models Learn Better on Perception Task](https://mlsanity.com/paper/2306.00693/) (2023)
- [DSHGT: Dual-Supervisors Heterogeneous Graph Transformer -- A pioneer study of using heterogeneous graph learning for detecting software vulnerabilities](https://mlsanity.com/paper/2306.01376/) (2023)
- [Fed MobiLLM: Efficient Federated LLM Fine-Tuning over Heterogeneous Mobile Devices via Server Assisted Side-Tuning](https://mlsanity.com/paper/2508.06765/) (2025)

## Built on by (most cited first, 1 of 1)

- [CHAMP: Cross-domain Hybrid Architecture for Matchmaking and Prediction in Online Multi-Player Games](https://mlsanity.com/paper/2609.04870/) (2026)

## Abstract (arXiv)

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.