Jul 7, 2022
When does Bias Transfer in Transfer Learning?
We demonstrate that biases from pre-trained models can persist even after fine-tuning.
Jun 30, 2022
Distilling Model Failures as Directions in Latent Space
We demonstrate how to distill patterns of model errors as directions in a latent space.
May 11, 2022
Uncovering Brittleness with Datamodels
In the second part of our datamodels series, we use datamodels to identify and study a new form of model brittleness.
Apr 20, 2022
Missingness Bias in Model Debugging
We demonstrate how current missingness approximations introduce biases into model debugging.
Feb 2, 2022
Predicting Predictions with Datamodels
In the first part of our datamodels series, we introduce datamodeling and its (linear) instantiation on CIFAR-10.
Dec 3, 2021
Editing a Classifier
We develop a methodology for directly editing the predictions rules of a pre-trained classifier with virtually no additional data collection.
Oct 17, 2021
Combining Diverse Feature Priors
We explore how a diverse set of feature priors can be leveraged to improve model generalization.
Oct 14, 2021
Certified Patch Robustness Via Smoothed Vision Transformers (Part 2)
We demonstrate how vision transformers lead to strong certified patch defenses with standard accuracy and inference times comparable to standard (non-robust) models.
