# How to Tune a Multilingual Encoder Model for Germanic Languages: A Study of PEFT, Full Fine-Tuning, and Language Adapters

arXiv: https://arxiv.org/abs/2501.06025 · HTML: https://mlsanity.com/paper/2501.06025/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 34 references that are in the map.
- First posted on arXiv 10 Jan 2025.

## Builds on (most cited first, 22 of 22)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing](https://mlsanity.com/paper/2111.09543/) (2021)
- [AdapterFusion: Non-Destructive Task Composition for Transfer Learning](https://mlsanity.com/paper/2005.00247/) (2020)
- [CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data](https://mlsanity.com/paper/1911.00359/) (2019)
- [MAD-X: An Adapter-Based Framework for Multi-Task Cross-Lingual Transfer](https://mlsanity.com/paper/2005.00052/) (2020)
- [On the Effectiveness of Adapter-based Tuning for Pretrained Language Model Adaptation](https://mlsanity.com/paper/2106.03164/) (2021)
- [Lifting the Curse of Multilinguality by Pre-training Modular Transformers](https://mlsanity.com/paper/2205.06266/) (2022)
- [Multilingual is not enough: BERT for Finnish](https://mlsanity.com/paper/1912.07076/) (2019)
- [Adapters: A Unified Library for Parameter-Efficient and Modular Transfer Learning](https://mlsanity.com/paper/2311.11077/) (2023)
- [GermanQuAD and GermanDPR: Improving Non-English Question Answering and Passage Retrieval](https://mlsanity.com/paper/2104.12741/) (2021)
- [Breaking the Curse of Multilinguality with Cross-lingual Expert Language Models](https://mlsanity.com/paper/2401.10440/) (2024)
- [Encoder vs Decoder: Comparative Analysis of Encoder and Decoder Language Models on Multilingual NLU Tasks](https://mlsanity.com/paper/2406.13469/) (2024)
- [Phylogeny-Inspired Adaptation of Multilingual Models to New Languages](https://mlsanity.com/paper/2205.09634/) (2022)
- [Orthogonal Language and Task Adapters in Zero-Shot Cross-Lingual Transfer](https://mlsanity.com/paper/2012.06460/) (2020)
- [Language-Family Adapters for Low-Resource Multilingual Neural Machine Translation](https://mlsanity.com/paper/2209.15236/) (2022)
- [A Warm Start and a Clean Crawled Corpus -- A Recipe for Good Language Models](https://mlsanity.com/paper/2201.05601/) (2022)
- [The Impact of Language Adapters in Cross-Lingual Transfer for NLU](https://mlsanity.com/paper/2402.00149/) (2024)
- [Modular Adaptation of Multilingual Encoders to Written Swiss German Dialect](https://mlsanity.com/paper/2401.14400/) (2024)

## Abstract (arXiv)

This paper investigates the optimal use of the multilingual encoder model mDeBERTa for tasks in three Germanic languages -- German, Swedish, and Icelandic -- representing varying levels of presence and likely data quality in mDeBERTas pre-training data. We compare full fine-tuning with the parameter-efficient fine-tuning (PEFT) methods LoRA and Pfeiffer bottleneck adapters, finding that PEFT is more effective for the higher-resource language, German. However, results for Swedish and Icelandic are less consistent. We also observe differences between tasks: While PEFT tends to work better for question answering, full fine-tuning is preferable for named entity recognition. Inspired by previous research on modular approaches that combine task and language adapters, we evaluate the impact of adding PEFT modules trained on unstructured text, finding that this approach is not beneficial.