# Unlocking the Potential of Model Merging for Low-Resource Languages

arXiv: https://arxiv.org/abs/2407.03994 · HTML: https://mlsanity.com/paper/2407.03994/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 47 references that are in the map.
- First posted on arXiv 4 Jul 2024.

## Builds on (most cited first, 30 of 33)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [TIES-Merging: Resolving Interference When Merging Models](https://mlsanity.com/paper/2306.01708/) (2023)
- [The Flan Collection: Designing Data and Methods for Effective Instruction Tuning](https://mlsanity.com/paper/2301.13688/) (2023)
- [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://mlsanity.com/paper/2311.03099/) (2023)
- [Merging Models with Fisher-Weighted Averaging](https://mlsanity.com/paper/2111.09832/) (2021)
- [Language Models are Multilingual Chain-of-Thought Reasoners](https://mlsanity.com/paper/2210.03057/) (2022)
- [MAD-X: An Adapter-Based Framework for Multi-Task Cross-Lingual Transfer](https://mlsanity.com/paper/2005.00052/) (2020)
- [Dataless Knowledge Fusion by Merging Weights of Language Models](https://mlsanity.com/paper/2212.09849/) (2022)
- [The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants](https://mlsanity.com/paper/2308.16884/) (2023)
- [Evolutionary Optimization of Model Merging Recipes](https://mlsanity.com/paper/2403.13187/) (2024)
- [Episodic Memory in Lifelong Language Learning](https://mlsanity.com/paper/1906.01076/) (2019)
- [Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning](https://mlsanity.com/paper/2402.06619/) (2024)
- [CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages](https://mlsanity.com/paper/2309.09400/) (2023)
- [SIB-200: A Simple, Inclusive, and Big Evaluation Dataset for Topic Classification in 200+ Languages and Dialects](https://mlsanity.com/paper/2309.07445/) (2023)
- [Understanding Catastrophic Forgetting in Language Models via Implicit Inference](https://mlsanity.com/paper/2309.10105/) (2023)
- [An Empirical Investigation of the Role of Pre-training in Lifelong Learning](https://mlsanity.com/paper/2112.09153/) (2021)
- [Fusing finetuned models for better pretraining](https://mlsanity.com/paper/2204.03044/) (2022)
- [SeaLLMs -- Large Language Models for Southeast Asia](https://mlsanity.com/paper/2312.00738/) (2023)
- [Towards Leaving No Indic Language Behind: Building Monolingual Corpora, Benchmark and Models for Indic Languages](https://mlsanity.com/paper/2212.05409/) (2022)
- [BLOOM+1: Adding Language Support to BLOOM for Zero-Shot Prompting](https://mlsanity.com/paper/2212.09535/) (2022)
- [Grid Search, Random Search, Genetic Algorithm: A Big Comparison for NAS](https://mlsanity.com/paper/1912.06059/) (2019)
- [Bactrian-X: Multilingual Replicable Instruction-Following Models with Low-Rank Adaptation](https://mlsanity.com/paper/2305.15011/) (2023)
- [Multilingual Instruction Tuning With Just a Pinch of Multilinguality](https://mlsanity.com/paper/2401.01854/) (2024)
- [AdaMergeX: Cross-Lingual Transfer with Large Language Models via Adaptive Adapter Merging](https://mlsanity.com/paper/2402.18913/) (2024)
- [InstructAlign: High-and-Low Resource Language Alignment via Continual Crosslingual Instruction Tuning](https://mlsanity.com/paper/2305.13627/) (2023)
- [X-Instruction: Aligning Language Model in Low-resource Languages with Self-curated Cross-lingual Instructions](https://mlsanity.com/paper/2405.19744/) (2024)

## Built on by (most cited first, 19 of 19)

- [Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities](https://mlsanity.com/paper/2408.07666/) (2024)
- [Mix Data or Merge Models? Optimizing for Diverse Multi-Task Learning](https://mlsanity.com/paper/2410.10801/) (2024)
- [Continually Adding New Languages to Multilingual Language Models](https://mlsanity.com/paper/2509.11414/) (2025)
- [Transport and Merge: Cross-Architecture Merging for Large Language Models](https://mlsanity.com/paper/2602.05495/) (2026)
- [Adapting Chat Language Models Using Only Target Unlabeled Language Data](https://mlsanity.com/paper/2412.11704/) (2024)
- [Why Do More Experts Fail? A Theoretical Analysis of Model Merging](https://mlsanity.com/paper/2505.21226/) (2025)
- [Fine-tuning Large Language Models with Limited Data: A Survey and Practical Guide](https://mlsanity.com/paper/2411.09539/) (2024)
- [Mitigating Catastrophic Forgetting in Target Language Adaptation of LLMs via Source-Shielded Updates](https://mlsanity.com/paper/2512.04844/) (2025)
- [Grow Up and Merge: Scaling Strategies for Efficient Language Adaptation](https://mlsanity.com/paper/2512.10772/) (2025)
- [Speech-FT: Merging Pre-trained And Fine-Tuned Speech Representation Models For Cross-Task Generalization](https://mlsanity.com/paper/2502.12672/) (2025)
- [One Model to Translate Them All? A Journey to Mount Doom for Multilingual Model Merging](https://mlsanity.com/paper/2604.02881/) (2026)
- [Multilinguality at the Edge: Developing Language Models for the Global South](https://mlsanity.com/paper/2604.21637/) (2026)
- [DiMtextsuperscript3: Bridging Multilingual and Multimodal Models via Direction- and Magnitude-Aware Merging](https://mlsanity.com/paper/2605.12960/) (2026)
- [A Pipeline to Assess Merging Methods via Behavior and Internals](https://mlsanity.com/paper/2509.19476/) (2025)
- [BYOL: Bring Your Own Language Into LLMs](https://mlsanity.com/paper/2601.10804/) (2026)
- [Mi:dm K 2.5 Pro](https://mlsanity.com/paper/2603.18788/) (2026)
- [Merge and Conquer: Instructing Multilingual Models by Adding Target Language Weights](https://mlsanity.com/paper/2603.28263/) (2026)
- [Efficient Low-Resource Language Adaptation via Multi-Source Dynamic Logit Fusion](https://mlsanity.com/paper/2604.18106/) (2026)
- [Towards Truly Multilingual ASR: Generalizing Code-Switching ASR to Unseen Language Pairs](https://mlsanity.com/paper/2606.05846/) (2026)

## Abstract (arXiv)

Adapting large language models (LLMs) to new languages typically involves continual pre-training (CT) followed by supervised fine-tuning (SFT). However, this CT-then-SFT approach struggles with limited data in the context of low-resource languages, failing to balance language modeling and task-solving capabilities. We thus propose model merging as an alternative for low-resource languages, combining models with distinct capabilities into a single model without additional training. We use model merging to develop task-solving LLMs for low-resource languages without SFT data in the target languages. Our experiments based on Llama-2-7B demonstrate that model merging effectively endows LLMs for low-resource languages with task-solving abilities, outperforming CT-then-SFT in scenarios with extremely scarce data. Observing performance saturation in model merging with more training tokens, we further analyze the merging process and introduce a slack variable to the model merging algorithm to mitigate the loss of important parameters, thereby enhancing performance. We hope that model merging can benefit more human languages suffering from data scarcity with its higher data efficiency.