Paper Detail

Sustainable Federated Fine-Tuning with LoRA: Environmental Impact and Performance Footprint

Pedro Augusto Londe Ribeiro, Rodrigo Moreira, Larissa Ferreira Rodrigues Moreira, Joberto S. B. Martins, Tereza C. Carvalho, Flávio de Oliveira Silva

arxiv Score 8.8

Published 2026-11-12 · First seen 2026-11-13

General AI

Abstract

Federated Learning (FL) with Parameter-Efficient Fine-Tuning (PEFT) enables efficient adaptation of Large Language Models (LLMs), yet its energy and carbon costs across hardware remain poorly understood. We evaluated federated Low-Rank Adaptation (LoRA) fine-tuning on NVIDIA A40, RTX 6000 Ada, and A100 GPUs using a $2^2$ full factorial design with 135M/360M models and 10/100 local steps. CodeCarbon tracked energy and estimated CO$_2$-equivalent emissions. Descriptive allocation of variation assigned 84.8-91.2% of energy and emissions variation to local steps within these ranges. On A40 and RTX 6000 Ada, configurations with fewer steps achieved comparable accuracy with roughly 6-8-fold lower energy consumption and higher accuracy per kWh. These findings support smaller local budgets for energy-efficient federated fine-tuning within the evaluated setting.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{ribeiro2026sustainable,
  title = {Sustainable Federated Fine-Tuning with LoRA: Environmental Impact and Performance Footprint},
  author = {Pedro Augusto Londe Ribeiro and Rodrigo Moreira and Larissa Ferreira Rodrigues Moreira and Joberto S. B. Martins and Tereza C. Carvalho and Flávio de Oliveira Silva},
  year = {2026},
  abstract = {Federated Learning (FL) with Parameter-Efficient Fine-Tuning (PEFT) enables efficient adaptation of Large Language Models (LLMs), yet its energy and carbon costs across hardware remain poorly understood. We evaluated federated Low-Rank Adaptation (LoRA) fine-tuning on NVIDIA A40, RTX 6000 Ada, and A100 GPUs using a \$2\textasciicircum{}2\$ full factorial design with 135M/360M models and 10/100 local steps. CodeCarbon tracked energy and estimated CO\$\_2\$-equivalent emissions. Descriptive allocation of variation assi},
  url = {https://arxiv.org/abs/2611.16856},
  keywords = {cs.LG, cs.NI},
  eprint = {2611.16856},
  archiveprefix = {arXiv},
}

Metadata

{}