# BREEDS: Benchmarks for Subpopulation Shift

arXiv: https://arxiv.org/abs/2008.04859 · HTML: https://mlsanity.com/paper/2008.04859/

## Key facts

- Cited by 155 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Aug 2020.

## Built on by (most cited first, 30 of 155)

- [SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks](https://mlsanity.com/paper/2310.03684/) (2023)
- [TRAK: Attributing Model Behavior at Scale](https://mlsanity.com/paper/2303.14186/) (2023)
- [Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalization](https://mlsanity.com/paper/2107.04649/) (2021)
- [Surgical Fine-Tuning Improves Adaptation to Distribution Shifts](https://mlsanity.com/paper/2210.11466/) (2022)
- [Leveraging Unlabeled Data to Predict Out-of-Distribution Performance](https://mlsanity.com/paper/2201.04234/) (2022)
- [Change is Hard: A Closer Look at Subpopulation Shift](https://mlsanity.com/paper/2302.12254/) (2023)
- [Model-Based Domain Generalization](https://mlsanity.com/paper/2102.11436/) (2021)
- [Plex: Towards Reliability using Pretrained Large Model Extensions](https://mlsanity.com/paper/2207.07411/) (2022)
- [Wild-Time: A Benchmark of in-the-Wild Distribution Shift over Time](https://mlsanity.com/paper/2211.14238/) (2022)
- [Distilling Model Failures as Directions in Latent Space](https://mlsanity.com/paper/2206.14754/) (2022)
- [MetaShift: A Dataset of Datasets for Evaluating Contextual Distribution Shifts and Training Conflicts](https://mlsanity.com/paper/2202.06523/) (2022)
- [CHiLS: Zero-Shot Image Classification with Hierarchical Label Sets](https://mlsanity.com/paper/2302.02551/) (2023)
- [Discover and Cure: Concept-aware Mitigation of Spurious Correlation](https://mlsanity.com/paper/2305.00650/) (2023)
- [Connect, Not Collapse: Explaining Contrastive Learning for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2204.00570/) (2022)
- [Mixture Outlier Exposure: Towards Out-of-Distribution Detection in Fine-grained Environments](https://mlsanity.com/paper/2106.03917/) (2021)
- [Probable Domain Generalization via Quantile Risk Minimization](https://mlsanity.com/paper/2207.09944/) (2022)
- [Editing a classifier by rewriting its prediction rules](https://mlsanity.com/paper/2112.01008/) (2021)
- [Text-To-Concept (and Back) via Cross-Model Alignment](https://mlsanity.com/paper/2305.06386/) (2023)
- [Contrastive Adapters for Foundation Model Group Robustness](https://mlsanity.com/paper/2207.07180/) (2022)
- [Beyond Model Adaptation at Test Time: A Survey](https://mlsanity.com/paper/2411.03687/) (2024)
- [A Call to Reflect on Evaluation Practices for Failure Detection in Image Classification](https://mlsanity.com/paper/2211.15259/) (2022)
- [Predicting Out-of-Distribution Error with the Projection Norm](https://mlsanity.com/paper/2202.05834/) (2022)
- [A Theory of Label Propagation for Subpopulation Shift](https://mlsanity.com/paper/2102.11203/) (2021)
- [Fusing Models with Complementary Expertise](https://mlsanity.com/paper/2310.01542/) (2023)
- [Beyond Separability: Analyzing the Linear Transferability of Contrastive Representations to Related Subpopulations](https://mlsanity.com/paper/2204.02683/) (2022)
- [ID and OOD Performance Are Sometimes Inversely Correlated on Real-world Datasets](https://mlsanity.com/paper/2209.00613/) (2022)
- [LLM as Dataset Analyst: Subpopulation Structure Discovery with Large Language Model](https://mlsanity.com/paper/2405.02363/) (2024)
- [Improving neural network representations using human similarity judgments](https://mlsanity.com/paper/2306.04507/) (2023)
- [Overcoming Common Flaws in the Evaluation of Selective Classification Systems](https://mlsanity.com/paper/2407.01032/) (2024)
- [Calibrated ensembles can mitigate accuracy tradeoffs under distribution shift](https://mlsanity.com/paper/2207.08977/) (2022)

## Abstract (arXiv)

We develop a methodology for assessing the robustness of models to subpopulation shift---specifically, their ability to generalize to novel data subpopulations that were not observed during training. Our approach leverages the class structure underlying existing datasets to control the data subpopulations that comprise the training and test distributions. This enables us to synthesize realistic distribution shifts whose sources can be precisely controlled and characterized, within existing large-scale datasets. Applying this methodology to the ImageNet dataset, we create a suite of subpopulation shift benchmarks of varying granularity. We then validate that the corresponding shifts are tractable by obtaining human baselines for them. Finally, we utilize these benchmarks to measure the sensitivity of standard model architectures as well as the effectiveness of off-the-shelf train-time robustness interventions. Code and data available at https://github.com/MadryLab/BREEDS-Benchmarks .