# Mix Data or Merge Models? Optimizing for Diverse Multi-Task Learning

arXiv: https://arxiv.org/abs/2410.10801 · HTML: https://mlsanity.com/paper/2410.10801/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 39 of its 46 references that are in the map.
- First posted on arXiv 14 Oct 2024.

## Builds on (most cited first, 30 of 39)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [Averaging Weights Leads to Wider Optima and Better Generalization](https://mlsanity.com/paper/1803.05407/) (2018)
- [Editing Models with Task Arithmetic](https://mlsanity.com/paper/2212.04089/) (2022)
- [Robustness May Be at Odds with Accuracy](https://mlsanity.com/paper/1805.12152/) (2018)
- [TIES-Merging: Resolving Interference When Merging Models](https://mlsanity.com/paper/2306.01708/) (2023)
- [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://mlsanity.com/paper/2311.03099/) (2023)
- [Linear Mode Connectivity and the Lottery Ticket Hypothesis](https://mlsanity.com/paper/1912.05671/) (2019)
- [Merging Models with Fisher-Weighted Averaging](https://mlsanity.com/paper/2111.09832/) (2021)
- [Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study](https://mlsanity.com/paper/2305.13860/) (2023)
- [Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions](https://mlsanity.com/paper/2309.07875/) (2023)
- [Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model](https://mlsanity.com/paper/2402.07827/) (2024)
- [Multilingual Jailbreak Challenges in Large Language Models](https://mlsanity.com/paper/2310.06474/) (2023)
- [Uniform convergence may be unable to explain generalization in deep learning](https://mlsanity.com/paper/1902.04742/) (2019)
- [Evolutionary Optimization of Model Merging Recipes](https://mlsanity.com/paper/2403.13187/) (2024)
- [Arcee's MergeKit: A Toolkit for Merging Large Language Models](https://mlsanity.com/paper/2403.13257/) (2024)
- [Gender bias and stereotypes in Large Language Models](https://mlsanity.com/paper/2308.14921/) (2023)
- [Model Breadcrumbs: Scaling Multi-Task Model Merging with Sparse Masks](https://mlsanity.com/paper/2312.06795/) (2023)
- [Knowledge Fusion of Large Language Models](https://mlsanity.com/paper/2401.10491/) (2024)
- [Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic](https://mlsanity.com/paper/2402.11746/) (2024)
- [The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts](https://mlsanity.com/paper/2401.13136/) (2024)
- [The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm](https://mlsanity.com/paper/2406.18682/) (2024)
- [Stochastic Weight Averaging in Parallel: Large-Batch Training that Generalizes Well](https://mlsanity.com/paper/2001.02312/) (2020)
- [All Languages Matter: On the Multilingual Safety of Large Language Models](https://mlsanity.com/paper/2310.00905/) (2023)
- [MetaGPT: Merging Large Language Models Using Model Exclusive Task Arithmetic](https://mlsanity.com/paper/2406.11385/) (2024)
- [LM-Cocktail: Resilient Tuning of Language Models via Model Merging](https://mlsanity.com/paper/2311.13534/) (2023)
- [Model Merging and Safety Alignment: One Bad Model Spoils the Bunch](https://mlsanity.com/paper/2406.14563/) (2024)
- [A safety realignment framework via subspace-oriented model fusion for large language models](https://mlsanity.com/paper/2405.09055/) (2024)

## Built on by (most cited first, 17 of 17)

- [MergeBench: A Benchmark for Merging Domain-Specialized LLMs](https://mlsanity.com/paper/2505.10833/) (2025)
- [Exact Unlearning of Finetuning Data via Model Merging at Scale](https://mlsanity.com/paper/2504.04626/) (2025)
- [OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging](https://mlsanity.com/paper/2505.19892/) (2025)
- [One Tokenizer To Rule Them All: Emergent Language Plasticity via Multilingual Tokenizers](https://mlsanity.com/paper/2506.10766/) (2025)
- [The Multilingual Divide and Its Impact on Global AI Safety](https://mlsanity.com/paper/2505.21344/) (2025)
- [When Life Gives You Samples: The Benefits of Scaling up Inference Compute for Multilingual LLMs](https://mlsanity.com/paper/2506.20544/) (2025)
- [Efficient Model Development through Fine-tuning Transfer](https://mlsanity.com/paper/2503.20110/) (2025)
- [The Art of Asking: Multilingual Prompt Optimization for Synthetic Data](https://mlsanity.com/paper/2510.19806/) (2025)
- [Toward a Holistic Approach to Continual Model Merging](https://mlsanity.com/paper/2509.23592/) (2025)
- [Decouple Searching from Training: Scaling Data Mixing via Model Merging for Large Language Model Pre-training](https://mlsanity.com/paper/2602.00747/) (2026)
- [GR-SAP: Generative Replay for Safety Alignment Preservation during Fine-Tuning](https://mlsanity.com/paper/2603.10243/) (2026)
- [Exploring Sparse Adapters for Scalable Merging of Parameter Efficient Experts](https://mlsanity.com/paper/2507.07140/) (2025)
- [BYOL: Bring Your Own Language Into LLMs](https://mlsanity.com/paper/2601.10804/) (2026)
- [Exploring Information Seeking Agent Consolidation](https://mlsanity.com/paper/2602.00585/) (2026)
- [Model Merging as Probabilistic Inference in Fine-Tuning Parameter Space](https://mlsanity.com/paper/2607.01689/) (2026)
- [Efficient Decentralized Multi-task Dataset Valuation via Model Merging](https://mlsanity.com/paper/2607.03346/) (2026)
- [Gaokerena: A Small Persian Medical Language Model Family](https://mlsanity.com/paper/2608.00932/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have been adopted and deployed worldwide for a broad variety of applications. However, ensuring their safe use remains a significant challenge. Preference training and safety measures often overfit to harms prevalent in Western-centric datasets, and safety protocols frequently fail to extend to multilingual settings. In this work, we explore model merging in a diverse multi-task setting, combining safety and general-purpose tasks within a multilingual context. Each language introduces unique and varied learning challenges across tasks. We find that objective-based merging is more effective than mixing data, with improvements of up to 8% and 10% in general performance and safety respectively. We also find that language-based merging is highly effective -- by merging monolingually fine-tuned models, we achieve a 4% increase in general performance and 7% reduction in harm across all languages on top of the data mixtures method using the same available data. Overall, our comprehensive study of merging approaches provides a useful framework for building strong and safe multilingual models.