# The Robustness Limits of SoTA Vision Models to Natural Variation

arXiv: https://arxiv.org/abs/2210.13604 · HTML: https://mlsanity.com/paper/2210.13604/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 46 references that are in the map.
- First posted on arXiv 24 Oct 2022.

## Builds on (most cited first, 30 of 34)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [iBOT: Image BERT Pre-Training with Online Tokenizer](https://mlsanity.com/paper/2111.07832/) (2021)
- [Disentangling by Factorising](https://mlsanity.com/paper/1802.05983/) (2018)
- [ImageNet-21K Pretraining for the Masses](https://mlsanity.com/paper/2104.10972/) (2021)
- [Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty](https://mlsanity.com/paper/1906.12340/) (2019)
- [Making Convolutional Networks Shift-Invariant Again](https://mlsanity.com/paper/1904.11486/) (2019)
- [ResNet strikes back: An improved training procedure in timm](https://mlsanity.com/paper/2110.00476/) (2021)
- [Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style](https://mlsanity.com/paper/2106.04619/) (2021)
- [An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution](https://mlsanity.com/paper/1807.03247/) (2018)
- [Understanding image representations by measuring their equivariance and equivalence](https://mlsanity.com/paper/1411.5908/) (2014)
- [Why do deep convolutional networks generalize so poorly to small image transformations?](https://mlsanity.com/paper/1805.12177/) (2018)
- [Self-supervised Pretraining of Visual Features in the Wild](https://mlsanity.com/paper/2103.01988/) (2021)
- [Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuning](https://mlsanity.com/paper/2003.12862/) (2020)
- [Strike (with) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects](https://mlsanity.com/paper/1811.11553/) (2018)
- [On Translation Invariance in CNNs: Convolutional Layers can Exploit Absolute Spatial Location](https://mlsanity.com/paper/2003.07064/) (2020)
- [On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Dataset](https://mlsanity.com/paper/1906.03292/) (2019)
- [Bridging Composite and Real: Towards End-to-end Deep Image Matting](https://mlsanity.com/paper/2010.16188/) (2020)
- [When Does Contrastive Learning Preserve Adversarial Robustness from Pretraining to Finetuning?](https://mlsanity.com/paper/2111.01124/) (2021)
- [High Fidelity Visualization of What Your Self-Supervised Representation Knows About](https://mlsanity.com/paper/2112.09164/) (2021)
- [A Framework For Contrastive Self-Supervised Learning And Designing A New Approach](https://mlsanity.com/paper/2009.00104/) (2020)
- [On the surprising similarities between supervised and self-supervised models](https://mlsanity.com/paper/2010.08377/) (2020)
- [Grounding inductive biases in natural images:invariance stems from variations in data](https://mlsanity.com/paper/2106.05121/) (2021)

## Built on by (most cited first, 15 of 15)

- [A Whac-A-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Others](https://mlsanity.com/paper/2212.04825/) (2022)
- [GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-distribution Generalization Perspective](https://mlsanity.com/paper/2211.08073/) (2022)
- [PUG: Photorealistic and Semantically Controllable Synthetic Data for Representation Learning](https://mlsanity.com/paper/2308.03977/) (2023)
- [A Closer Look at Benchmarking Self-Supervised Pre-training with Image Classification](https://mlsanity.com/paper/2407.12210/) (2024)
- [Does Progress On Object Recognition Benchmarks Improve Real-World Generalization?](https://mlsanity.com/paper/2307.13136/) (2023)
- [Are Deep Neural Networks Adequate Behavioural Models of Human Visual Perception?](https://mlsanity.com/paper/2305.17023/) (2023)
- [On the Ability of Deep Networks to Learn Symmetries from Data: A Neural Kernel Theory](https://mlsanity.com/paper/2412.11521/) (2024)
- [A comparison between humans and AI at recognizing objects in unusual poses](https://mlsanity.com/paper/2402.03973/) (2024)
- [ObjectCompose: Evaluating Resilience of Vision-Based Models on Object-to-Background Compositional Changes](https://mlsanity.com/paper/2403.04701/) (2024)
- [GeSS: Benchmarking Geometric Deep Learning under Scientific Applications with Distribution Shifts](https://mlsanity.com/paper/2310.08677/) (2023)
- [Causality-Driven Audits of Model Robustness](https://mlsanity.com/paper/2410.23494/) (2024)
- [MoENAS: Mixture-of-Expert based Neural Architecture Search for jointly Accurate, Fair, and Robust Edge Deep Neural Networks](https://mlsanity.com/paper/2502.07422/) (2025)
- [Self-Supervised Learning from Structural Invariance](https://mlsanity.com/paper/2602.02381/) (2026)
- [A Causal Framework for Aligning Image Quality Metrics and Deep Neural Network Robustness](https://mlsanity.com/paper/2503.02797/) (2025)
- [Latent Equivariant Operators for Robust Object Recognition: Promises and Challenges](https://mlsanity.com/paper/2602.18406/) (2026)

## Abstract (arXiv)

Recent state-of-the-art vision models introduced new architectures, learning paradigms, and larger pretraining data, leading to impressive performance on tasks such as classification. While previous generations of vision models were shown to lack robustness to factors such as pose, it's unclear the extent to which this next generation of models are more robust. To study this question, we develop a dataset of more than 7 million images with controlled changes in pose, position, background, lighting, and size. We study not only how robust recent state-of-the-art models are, but also the extent to which models can generalize variation in factors when they're present during training. We consider a catalog of recent vision models, including vision transformers (ViT), self-supervised models such as masked autoencoders (MAE), and models trained on larger datasets such as CLIP. We find out-of-the-box, even today's best models are not robust to common changes in pose, size, and background. When some samples varied during training, we found models required a significant portion of diversity to generalize -- though eventually robustness did improve. When diversity is only seen for some classes however, we found models did not generalize to other classes, unless the classes were very similar to those seen varying during training. We hope our work will shed further light on the blind spots of SoTA models and spur the development of more robust vision models.