Skip to content
#federated learning Open access

ATLAS: Adaptive Task-Aware Federated Learning with Heterogeneous LoRA and Split Learning for Edge LLMs

Nov 2026 · Proceedings of the AAAI Symposium Series · 0 citations
Privacy-Preserving Technologies in Data

Abstract

Federated learning (FL) can adapt large language models (LLMs) without centralizing private data, but deployment on heterogeneous edge clients is constrained by task diversity, device variability, and communication cost. Standard task-agnostic aggregation and uniform-rank low-rank adaptation (LoRA) often cause negative transfer and poor resource matching. This paper presents ATLAS, a task-aware framework for parameter-efficient federated fine-tuning of transformer LLMs on heterogeneous clients. ATLAS combines representation-based task discovery, budget-aware heterogeneous LoRA allocation, state-consistent split federated learning, and LoRA-only graph regularization, together with INT8 LoRA communication and FP32 server-side computation. On a heterogeneous GLUE federation with three tasks, nine clients, and three backbones, ATLAS improves the macroaverage task score over standard FedAvg by 16.18 points on DistilBERT, 24.83 points on GPT-2, and 6.86 points on Qwen-2.5-0.5B, while reducing LoRA synchronization traffic and eliminating low-resource-client failures.

Read PDF

Similar papers

#machine learning Review Open access Oct 2014

Software development in startup companies: A systematic mapping study

The results indicate that software engineering work practices are chosen opportunistically, adapted and configured to provide value under the constrains imposed by the startup context.

Nicolò Paternoster, Carmine Giardino, M. Unterkalmsteiner et al. · 394 citations · ⚡54
#machine learning Review Open access Jun 2014

Why Early-Stage Software Startups Fail: A Behavioral Framework

This state-of-practice investigation was performed using a literature review followed by a multiple-case study approach and presents how inconsistency between managerial strategies and execution can lead to failure by means of a behavioral framework.

Carmine Giardino, Xiaofeng Wang, P. Abrahamsson · 175 citations · ⚡19
#machine learning Review Open access Oct 2016

“Failures” to be celebrated: an analysis of major pivots of software startups

This study conducts a case survey study based on the secondary data of the major pivots happened in 49 software startups, and demonstrates that customer need pivot is the most common among all pivot types.

Sohaib Shahid Bajwa, Xiaofeng Wang, Anh Nguyen-Duc et al. · 127 citations · ⚡15
#machine learning Review Open access May 2016

Key Challenges in Software Startups Across Life Cycle Stages

It is found that what perceived as biggest challenges by software startups do vary across different life cycle stages, even though its significance decreases when the learning focuses of the startups move from problem to solution and their products mature.

Xiaofeng Wang, Henry Edison, Sohaib Shahid Bajwa et al. · 62 citations · ⚡6

Related blog posts

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.