# Is Large-Scale Pretraining the Secret to Good Domain Generalization?

arXiv: https://arxiv.org/abs/2412.02856 · HTML: https://mlsanity.com/paper/2412.02856/

## Key facts

- Cited by 9 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 3 Dec 2024.

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

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Moment Matching for Multi-Source Domain Adaptation](https://mlsanity.com/paper/1812.01754/) (2018)
- [Adversarial Discriminative Domain Adaptation](https://mlsanity.com/paper/1702.05464/) (2017)
- [Averaging Weights Leads to Wider Optima and Better Generalization](https://mlsanity.com/paper/1803.05407/) (2018)
- [Deep CORAL: Correlation Alignment for Deep Domain Adaptation](https://mlsanity.com/paper/1607.01719/) (2016)
- [In Search of Lost Domain Generalization](https://mlsanity.com/paper/2007.01434/) (2020)
- [Learning Robust Global Representations by Penalizing Local Predictive Power](https://mlsanity.com/paper/1905.13549/) (2019)
- [Deeper, Broader and Artier Domain Generalization](https://mlsanity.com/paper/1710.03077/) (2017)
- [Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution](https://mlsanity.com/paper/2202.10054/) (2022)
- [Recognition in Terra Incognita](https://mlsanity.com/paper/1807.04975/) (2018)
- [SWAD: Domain Generalization by Seeking Flat Minima](https://mlsanity.com/paper/2102.08604/) (2021)
- [Scaling Open-Vocabulary Object Detection](https://mlsanity.com/paper/2306.09683/) (2023)
- [Explicit Inductive Bias for Transfer Learning with Convolutional Networks](https://mlsanity.com/paper/1802.01483/) (2018)
- [Improve Unsupervised Domain Adaptation with Mixup Training](https://mlsanity.com/paper/2001.00677/) (2020)
- [Diverse Weight Averaging for Out-of-Distribution Generalization](https://mlsanity.com/paper/2205.09739/) (2022)
- [Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalization](https://mlsanity.com/paper/2110.10832/) (2021)
- [Data Determines Distributional Robustness in Contrastive Language Image Pre-training (CLIP)](https://mlsanity.com/paper/2205.01397/) (2022)
- [Domain Generalization by Mutual-Information Regularization with Pre-trained Models](https://mlsanity.com/paper/2203.10789/) (2022)
- [Revisiting Weakly Supervised Pre-Training of Visual Perception Models](https://mlsanity.com/paper/2201.08371/) (2022)
- [Sharpness-Aware Gradient Matching for Domain Generalization](https://mlsanity.com/paper/2303.10353/) (2023)
- [CLIPood: Generalizing CLIP to Out-of-Distributions](https://mlsanity.com/paper/2302.00864/) (2023)
- [Three things everyone should know about Vision Transformers](https://mlsanity.com/paper/2203.09795/) (2022)
- [No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance](https://mlsanity.com/paper/2404.04125/) (2024)
- [Adversarial Style Augmentation for Domain Generalized Urban-Scene Segmentation](https://mlsanity.com/paper/2207.04892/) (2022)

## Built on by (most cited first, 9 of 9)

- [What's in a Latent? Leveraging Diffusion Latent Space for Domain Generalization](https://mlsanity.com/paper/2503.06698/) (2025)
- [Noise-Aware Generalization: Robustness to In-Domain Noise and Out-of-Domain Generalization](https://mlsanity.com/paper/2504.02996/) (2025)
- [Out-of-Distribution Generalization in the ARC-AGI Domain: Comparing Execution-Guided Neural Program Synthesis and Test-Time Fine-Tuning](https://mlsanity.com/paper/2507.15877/) (2025)
- [Domain Generalization in-the-Wild: Disentangling Classification from Domain-Aware Representations](https://mlsanity.com/paper/2508.21769/) (2025)
- [A General-Purpose Diversified 2D Seismic Image Dataset from NAMSS](https://mlsanity.com/paper/2602.04890/) (2026)
- [MaRS: Robust Out-of-Distribution Detection via Mahalanobis Residual Scoring](https://mlsanity.com/paper/2606.22649/) (2026)
- [Vertical Fusion: Condensing Internal Representations for Robust ViT Classification](https://mlsanity.com/paper/2607.10391/) (2026)
- [Domain Generalization for Smartphone-Based Human Activity Recognition: A Systematic Analysis of Components and Interactions](https://mlsanity.com/paper/2609.14863/) (2026)
- [PhaseAT: Fourier Phase Adversarial Training for Medical Image Domain Generalization](https://mlsanity.com/paper/2610.01807/) (2026)

## Abstract (arXiv)

Multi-Source Domain Generalization (DG) is the task of training on multiple source domains and achieving high classification performance on unseen target domains. Recent methods combine robust features from web-scale pretrained backbones with new features learned from source data, and this has dramatically improved benchmark results. However, it remains unclear if DG finetuning methods are becoming better over time, or if improved benchmark performance is simply an artifact of stronger pre-training. Prior studies have shown that perceptual similarity to pre-training data correlates with zero-shot performance, but we find the effect limited in the DG setting. Instead, we posit that having perceptually similar data in pretraining is not enough; and that it is how well these data were learned that determines performance. This leads us to introduce the Alignment Hypothesis, which states that the final DG performance will be high if and only if alignment of image and class label text embeddings is high. Our experiments confirm the Alignment Hypothesis is true, and we use it as an analysis tool of existing DG methods evaluated on DomainBed datasets by splitting evaluation data into In-pretraining (IP) and Out-of-pretraining (OOP). We show that all evaluated DG methods struggle on DomainBed-OOP, while recent methods excel on DomainBed-IP. Put together, our findings highlight the need for DG methods which can generalize beyond pretraining alignment.