Neural network classifiers can largely rely on simple spurious features, such as
backgrounds, to make predictions. However, even in these cases, we show that they still …

Given a language model (LM), maximum probability is a poor decoding objective for open-
ended generation, because it produces short and repetitive text. On the other hand …

Learning models that gracefully handle distribution shifts is central to research on domain
generalization, robust optimization, and fairness. A promising formulation is domain …

We study the problem of learning classifiers that perform well across (known or unknown)
groups of data. After observing that common worst-group-accuracy datasets suffer from …

Deep neural networks (DNNs) have had extraordinary successes in classifying
photographic images of objects and are often described as the best models of biological …

Much recent work in NLP has documented dataset artifacts, bias, and spurious correlations
between input features and output labels. However, how to tell which features have" …

The trustworthiness of machine learning has emerged as a critical topic in the field,
encompassing various applications and research areas such as robustness, security …

State-of-the-art stereo matching networks trained only on synthetic data often fail to
generalize to more challenging real data domains. In this paper, we attempt to unfold an …

Neural networks trained with ERM (empirical risk minimization) sometimes learn unintended
decision rules, in particular when their training data is biased, ie, when training labels are …

We consider the problem of training a classification model with group annotated training
data. Recent work has established that, if there is distribution shift across different groups …