# Demystifying Training-Time Augmentation for Data-Constrained Language Model Pretraining

arXiv: https://arxiv.org/abs/2606.16246 · HTML: https://mlsanity.com/paper/2606.16246/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 49 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Builds on (most cited first, 30 of 32)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research](https://mlsanity.com/paper/2402.00159/) (2024)
- [DataComp-LM: In search of the next generation of training sets for language models](https://mlsanity.com/paper/2406.11794/) (2024)
- [Efficient Training of Language Models to Fill in the Middle](https://mlsanity.com/paper/2207.14255/) (2022)
- [Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations](https://mlsanity.com/paper/2405.18392/) (2024)
- [Nemotron-CC: Transforming Common Crawl into a Refined Long-Horizon Pretraining Dataset](https://mlsanity.com/paper/2412.02595/) (2024)
- [Scaling Laws and Interpretability of Learning from Repeated Data](https://mlsanity.com/paper/2205.10487/) (2022)
- [Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learning](https://mlsanity.com/paper/2303.02861/) (2023)
- [To Repeat or Not To Repeat: Insights from Scaling LLM under Token-Crisis](https://mlsanity.com/paper/2305.13230/) (2023)
- [Diffusion Beats Autoregressive in Data-Constrained Settings](https://mlsanity.com/paper/2507.15857/) (2025)
- [Diffusion Language Models are Super Data Learners](https://mlsanity.com/paper/2511.03276/) (2025)
- [Theoretical Benefit and Limitation of Diffusion Language Model](https://mlsanity.com/paper/2502.09622/) (2025)
- [Coherence boosting: When your pretrained language model is not paying enough attention](https://mlsanity.com/paper/2110.08294/) (2021)
- [Pre-training under infinite compute](https://mlsanity.com/paper/2509.14786/) (2025)
- [Self-MoE: Towards Compositional Large Language Models with Self-Specialized Experts](https://mlsanity.com/paper/2406.12034/) (2024)
- [Meet in the Middle: A New Pre-training Paradigm](https://mlsanity.com/paper/2303.07295/) (2023)
- [DeepPersona: A Generative Engine for Scaling Deep Synthetic Personas](https://mlsanity.com/paper/2511.07338/) (2025)

## Built on by (most cited first, 4 of 4)

- [The Announcement Carries the Cue: Markup, Boundaries, and the Notation of Pre-Training Corpora](https://mlsanity.com/paper/2608.09093/) (2026)
- [Decoupling Token Roles in Autoregressive Pretraining](https://mlsanity.com/paper/2609.33405/) (2026)
- [What Shared Prefixes Hide: Trajectory Dropout for On-Policy Distillation](https://mlsanity.com/paper/2609.33455/) (2026)
- [What Is a Repeated Token Worth? The Scaling Geometry of Multi-Epoch Pretraining](https://mlsanity.com/paper/2610.05591/) (2026)

## Abstract (arXiv)

As AI labs approach a data ceiling where compute capacity outpaces the rate of new high-quality text generation, language model pretraining is shifting toward a data-constrained, compute-abundant regime that demands productive multi-epoch training on fixed corpora. Standard autoregressive (AR) pretraining overfits severely in this setting, reaching its optimum early and then continuously deteriorating. We investigate training-time data augmentation as a regularizer to mitigate this overfitting and enable productive training for hundreds of epochs on the same data. We introduce three orthogonal categories of augmentation for AR pretraining: token-level noise (masking, random replacement), sequence permutations (right-to-left prediction, Fill-in-the-Middle), and target offset prediction (xt+i for i > 1). Through systematic ablations, we find that individual augmentations delay overfitting and lower validation loss relative to the baseline, with random token replacement achieving the best minimum loss among individual methods. Combining augmentation categories further lowers the minimum validation loss. Our experiments demonstrate that data augmentations mitigate AR pretraining's data inefficiency and offer a promising solution to the data-constrained regime~footnoteAll code and data are available at https://github.com/ michaelchen-lab/ data-augmentations-for-pretraining.