# Early Data Exposure Improves Robustness to Subsequent Fine-Tuning

arXiv: https://arxiv.org/abs/2605.12705 · HTML: https://mlsanity.com/paper/2605.12705/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 28 references that are in the map.
- First posted on arXiv 12 May 2026.

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

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Robust fine-tuning of zero-shot models](https://mlsanity.com/paper/2109.01903/) (2021)
- [TOFU: A Task of Fictitious Unlearning for LLMs](https://mlsanity.com/paper/2401.06121/) (2024)
- [SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model](https://mlsanity.com/paper/2502.02737/) (2025)
- [Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models](https://mlsanity.com/paper/2310.02949/) (2023)
- [Implicit Regularization of Discrete Gradient Dynamics in Linear Neural Networks](https://mlsanity.com/paper/1904.13262/) (2019)
- [Understanding Emergent Abilities of Language Models from the Loss Perspective](https://mlsanity.com/paper/2403.15796/) (2024)
- [Overtrained Language Models Are Harder to Fine-Tune](https://mlsanity.com/paper/2503.19206/) (2025)
- [Scaling Laws for Predicting Downstream Performance in LLMs](https://mlsanity.com/paper/2410.08527/) (2024)
- [Safety Pretraining: Toward the Next Generation of Safe AI](https://mlsanity.com/paper/2504.16980/) (2025)
- [Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs](https://mlsanity.com/paper/2508.06601/) (2025)
- [Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection](https://mlsanity.com/paper/2502.06042/) (2025)
- [Representation Shattering in Transformers: A Synthetic Study with Knowledge Editing](https://mlsanity.com/paper/2410.17194/) (2024)
- [Replaying pre-training data improves fine-tuning](https://mlsanity.com/paper/2603.04964/) (2026)
- [The Finetuner's Fallacy: When to Pretrain with Your Finetuning Data](https://mlsanity.com/paper/2603.16177/) (2026)
- [When Should We Introduce Safety Interventions During Pretraining?](https://mlsanity.com/paper/2601.07087/) (2026)

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

- [DataComp-VLM: Improved Open Datasets for Vision-Language Models](https://mlsanity.com/paper/2606.28551/) (2026)
- [Synthetic Persona Pretraining: Alignment from Token Zero](https://mlsanity.com/paper/2608.13482/) (2026)
- [Similar Models Learn Differently: Final-Window Pretraining Shapes Post-Training Beyond SFT](https://mlsanity.com/paper/2607.25063/) (2026)
- [Fine Until Fine-Tuned: Repeated Solutions Make Reasoning Fragile](https://mlsanity.com/paper/2609.33559/) (2026)

## Abstract (arXiv)

How can we train models whose post-trained capabilities survive subsequent fine-tuning? Rather than focusing on downstream interventions to mitigate forgetting of upstream capabilities, we study how upstream training choices - that is, the manner in which a capability is acquired - shape how robustly that capability is retained. We investigate this question in a controlled three-stage language-model pipeline: pretraining, post-training to acquire a target capability, and downstream fine-tuning on a new objective. Across 135M and 1B models, two post-training domains, and two downstream fine-tuning tasks, we find that immediate post-training performance does not reliably predict retention after subsequent fine-tuning: training recipes that look equivalent immediately after post-training can retain the target capability very differently after subsequent fine-tuning. In particular, early exposure - mixing post-training data into pretraining - consistently improves the frontier between retained upstream performance and downstream performance. In compute-matched experiments, where the target data must be allocated between pretraining and post-training, we find that the optimum lies at neither extreme. Together with our other empirical and theoretical findings, this supports the view that post-training drives immediate specialization while early exposure improves robustness to later forgetting. Replay and dropout, typically used to mitigate forgetting as it occurs during fine-tuning, provide complementary gains to early exposure when applied during post-training. Our findings suggest that robustness to subsequent fine-tuning should be treated as a first-class objective of upstream training, addressed preventatively through choices like early exposure rather than reactively during fine-tuning itself.