# Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data

arXiv: https://arxiv.org/abs/2609.11917 · HTML: https://mlsanity.com/paper/2609.11917/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in MoE large language models, one of 12 directions in Mixture-of-experts.
- Builds on 21 of its 32 references that are in the map.
- First posted on arXiv 10 Sep 2026.

## Placed in

- Mixture-of-experts / [MoE large language models](https://mlsanity.com/moe/moe-large-language-models/)

## Builds on (most cited first, 21 of 21)

- [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)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus](https://mlsanity.com/paper/2104.08758/) (2021)
- [Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research](https://mlsanity.com/paper/2402.00159/) (2024)
- [Scaling Data-Constrained Language Models](https://mlsanity.com/paper/2305.16264/) (2023)
- [ST-MoE: Designing Stable and Transferable Sparse Expert Models](https://mlsanity.com/paper/2202.08906/) (2022)
- [The Stack: 3 TB of permissively licensed source code](https://mlsanity.com/paper/2211.15533/) (2022)
- [Scaling Laws and Interpretability of Learning from Repeated Data](https://mlsanity.com/paper/2205.10487/) (2022)
- [To Repeat or Not To Repeat: Insights from Scaling LLM under Token-Crisis](https://mlsanity.com/paper/2305.13230/) (2023)
- [M2D2: A Massively Multi-domain Language Modeling Dataset](https://mlsanity.com/paper/2210.07370/) (2022)
- [Prescriptive Scaling Laws for Data Constrained Training](https://mlsanity.com/paper/2605.01640/) (2026)
- [Datasets, Documents, and Repetitions: The Practicalities of Unequal Data Quality](https://mlsanity.com/paper/2503.07879/) (2025)
- [A Bitter Lesson for Data Filtering](https://mlsanity.com/paper/2605.19407/) (2026)
- [Scale Dependent Data Duplication](https://mlsanity.com/paper/2603.06603/) (2026)
- [Internal Data Repetition Destroys Language Models](https://mlsanity.com/paper/2606.24998/) (2026)

## Built on by (most cited first, 2 of 2)

- [Forking: Sudden Overfitting Under Replay](https://mlsanity.com/paper/2610.00394/) (2026)
- [Scaling Laws for Multimodal Data Mixtures](https://mlsanity.com/paper/2610.14423/) (2026)

## Abstract (arXiv)

As the supply of human-written text is exhausted, it has become standard practice to repeat language model training data. Prior work has studied data repetition for densely activated Transformers, but the effects of data repetition remains largely unexplored for recently dominant sparse architectures such as Mixture-of-Experts (MoE), despite their increased compute efficiency. We vary data repetition rates across single- and multi-domain data mixes, and across MoE settings, including expert count and granularity. We consistently find, for models ranging from 80M to 1B active (8.5B total) parameters, that MoEs degrade more rapidly under data repetition. This effect increases with sparsity, dictated by total rather than active parameters. While 80M dense models can repeat data over 8x with minimal degradation, MoEs instead begin to suffer at 4x, and deteriorate rapidly, ceding their performance benefits in all-unique data settings to underperform dense models after 32x. We experiment with existing regularization methods as a potential remedy. We find that some methods, such as dropout, can mitigate overfitting. In particular, with strong masking-based regularization, MoEs are able to outperform dense models even when data is repeated more than 64 times. However, no method fully matches the performance of all-unique training data. Finally, we analyze internal mechanisms correlated with MoE overfitting in high repetition regimes, and find that MoE routing universally stabilizes early in training, and that expert specialization correlates with overfitting to repeated data. In sum, our work addresses the adverse interactions between sparsity and data repetition: we present evidence for the core mechanisms of overfitting and its potential remediation, and suggest promising avenues for future methods to reduce over-specialization in model parameters by disrupting memorization patterns.