# Learning to Collaborate: An Orchestrated-Decentralized Framework for Peer-to-Peer LLM Federation

arXiv: https://arxiv.org/abs/2601.17133 · HTML: https://mlsanity.com/paper/2601.17133/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 34 references that are in the map.
- First posted on arXiv 23 Jan 2026.

## Builds on (most cited first, 16 of 16)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Federated Optimization in Heterogeneous Networks](https://mlsanity.com/paper/1812.06127/) (2018)
- [Inverting Gradients -- How easy is it to break privacy in federated learning?](https://mlsanity.com/paper/2003.14053/) (2020)
- [CodeBLEU: a Method for Automatic Evaluation of Code Synthesis](https://mlsanity.com/paper/2009.10297/) (2020)
- [FedMD: Heterogenous Federated Learning via Model Distillation](https://mlsanity.com/paper/1910.03581/) (2019)
- [BrainTorrent: A Peer-to-Peer Environment for Decentralized Federated Learning](https://mlsanity.com/paper/1905.06731/) (2019)
- [Federated Learning of Large Language Models with Parameter-Efficient Prompt Tuning and Adaptive Optimization](https://mlsanity.com/paper/2310.15080/) (2023)
- [FedBalancer: Data and Pace Control for Efficient Federated Learning on Heterogeneous Clients](https://mlsanity.com/paper/2201.01601/) (2022)
- [Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices](https://mlsanity.com/paper/2103.03239/) (2021)
- [IPLS : A Framework for Decentralized Federated Learning](https://mlsanity.com/paper/2101.01901/) (2021)
- [Personalized Decentralized Federated Learning with Knowledge Distillation](https://mlsanity.com/paper/2302.12156/) (2023)
- [Coral Protocol: Open Infrastructure Connecting The Internet of Agents](https://mlsanity.com/paper/2505.00749/) (2025)
- [Evaluating the Cybersecurity Risk of Real World, Machine Learning Production Systems](https://mlsanity.com/paper/2107.01806/) (2021)

## Built on by (most cited first, 2 of 2)

- [Wireless Federated Multi-Task LLM Fine-Tuning via Sparse-and-Orthogonal LoRA](https://mlsanity.com/paper/2602.20492/) (2026)
- [Priority-Aware Learning-Unlearning Correction for Dynamic Decentralized LoRA Fine-Tuning](https://mlsanity.com/paper/2606.22878/) (2026)

## Abstract (arXiv)

Fine-tuning Large Language Models (LLMs) for specialized domains is constrained by a fundamental challenge: the need for diverse, cross-organizational data conflicts with the principles of data privacy and sovereignty. While Federated Learning (FL) provides a framework for collaboration without raw data exchange, its classic centralized form introduces a single point of failure and remains vulnerable to model inversion attacks. Decentralized FL (DFL) mitigates this risk by removing the central aggregator but typically relies on inefficient, random peer-to-peer (P2P) pairings, forming a collaboration graph that is blind to agent heterogeneity and risks negative transfer. This paper introduces KNEXA-FL, a novel framework for orchestrated decentralization that resolves this trade-off. KNEXA-FL employs a non-aggregating Central Profiler/Matchmaker (CPM) that formulates P2P collaboration as a contextual bandit problem, using a LinUCB algorithm on abstract agent profiles to learn an optimal matchmaking policy. It orchestrates direct knowledge exchange between heterogeneous, PEFT-based LLM agents via secure distillation, without ever accessing the models themselves. Our comprehensive experiments on a challenging code generation task show that KNEXA-FL yields substantial gains, improving Pass@1 by approx. 50% relative to random P2P collaboration. Critically, our orchestrated approach demonstrates stable convergence, in stark contrast to a powerful centralized distillation baseline which suffers from catastrophic performance collapse. Our work establishes adaptive, learning-based orchestration as a foundational principle for building robust and effective decentralized AI ecosystems.