Detail publikačního výsledku

PEFT-Factory: Unified Parameter-Efficient Fine-Tuning of Autoregressive Large Language Models

BELANEC, R.; SRBA, I.; BIELIKOVÁ, M.

Originální název

PEFT-Factory: Unified Parameter-Efficient Fine-Tuning of Autoregressive Large Language Models

Anglický název

PEFT-Factory: Unified Parameter-Efficient Fine-Tuning of Autoregressive Large Language Models

Druh

Stať ve sborníku mimo WoS a Scopus

Originální abstrakt

Parameter-Efficient Fine-Tuning (PEFT) methods address the increasing size of Large Language Models (LLMs). Currently, many newly introduced PEFT methods are challenging to replicate, deploy, or compare with one another. To address this, we introduce PEFT-Factory, a unified framework for efficient fine-tuning LLMs using both off-the-shelf and custom PEFT methods. While its modular design supports extensibility, it natively provides a representative set of 19 PEFT methods, 27 classification and text generation datasets addressing 12 tasks, and both standard and PEFT-specific evaluation metrics. As a result, PEFT-Factory provides a ready-to-use, controlled, and stable environment, improving replicability and benchmarking of PEFT methods. PEFT-Factory is a downstream framework that originates from the popular LLaMA-Factory, and is publicly available at https://github.com/kinit-sk/PEFT-Factory.

Anglický abstrakt

Parameter-Efficient Fine-Tuning (PEFT) methods address the increasing size of Large Language Models (LLMs). Currently, many newly introduced PEFT methods are challenging to replicate, deploy, or compare with one another. To address this, we introduce PEFT-Factory, a unified framework for efficient fine-tuning LLMs using both off-the-shelf and custom PEFT methods. While its modular design supports extensibility, it natively provides a representative set of 19 PEFT methods, 27 classification and text generation datasets addressing 12 tasks, and both standard and PEFT-specific evaluation metrics. As a result, PEFT-Factory provides a ready-to-use, controlled, and stable environment, improving replicability and benchmarking of PEFT methods. PEFT-Factory is a downstream framework that originates from the popular LLaMA-Factory, and is publicly available at https://github.com/kinit-sk/PEFT-Factory.

Klíčová slova

Parameter-Efficient Fine-Tuning, Large Language Models, Autoregressive Models, Natural Language Processing

Klíčová slova v angličtině

Parameter-Efficient Fine-Tuning, Large Language Models, Autoregressive Models, Natural Language Processing

Autoři

BELANEC, R.; SRBA, I.; BIELIKOVÁ, M.

Vydáno

01.01.2026

Nakladatel

Association for Computational Linguistics

Místo

Morocco

ISBN

979-8-89176-382-1

Kniha

Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations)

Strany od

188

Strany do

202

Strany počet

15

URL

https://aclanthology.org/2026.eacl-demo.15/

BibTex

@inproceedings{BUT201831,
  author="{} and Róbert {Belanec} and  {} and Ivan {Srba} and  {} and Mária {Bieliková}",
  title="PEFT-Factory: Unified Parameter-Efficient Fine-Tuning of Autoregressive Large Language Models",
  booktitle="Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations)",
  year="2026",
  pages="188--202",
  publisher="Association for Computational Linguistics",
  address="Morocco",
  doi="10.18653/v1/2026.eacl-demo.15",
  isbn="979-8-89176-382-1",
  url="https://aclanthology.org/2026.eacl-demo.15/"
}