On the need for a language describing distribution shifts: Illustrations on tabular datasets

J Liu, T Wang, P Cui… - Advances in Neural …, 2024‏ - proceedings.neurips.cc
Different distribution shifts require different algorithmic and operational interventions.
Methodological research must be grounded by the specific shifts they address. Although …

Demographic bias in misdiagnosis by computational pathology models

A Vaidya, RJ Chen, DFK Williamson, AH Song… - Nature Medicine, 2024‏ - nature.com
Despite increasing numbers of regulatory approvals, deep learning-based computational
pathology systems often overlook the impact of demographic factors on performance …

A survey on evaluation of out-of-distribution generalization

H Yu, J Liu, X Zhang, J Wu, P Cui - arxiv preprint arxiv:2403.01874, 2024‏ - arxiv.org
Machine learning models, while progressively advanced, rely heavily on the IID assumption,
which is often unfulfilled in practice due to inevitable distribution shifts. This renders them …

The data addition dilemma

JH Shen, ID Raji, IY Chen - arxiv preprint arxiv:2408.04154, 2024‏ - arxiv.org
In many machine learning for healthcare tasks, standard datasets are constructed by
amassing data across many, often fundamentally dissimilar, sources. But when does adding …

Doubly robust augmented model accuracy transfer inference with high dimensional features

D Zhou, M Liu, M Li, T Cai - Journal of the American Statistical …, 2024‏ - Taylor & Francis
Transfer learning is crucial for training models that generalize to unlabeled target
populations using labeled source data, especially in real-world studies where label scarcity …

AdaptSel: Adaptive Selection of Biased and Debiased Recommendation Models for Varying Test Environments

Z Wang, H Zou, J Liu, J Wu, P Tian, Y He… - ACM Transactions on …, 2025‏ - dl.acm.org
Recommendation systems are frequently challenged by pervasive biases in the training set
that can compromise model effectiveness. To address this issue, various debiasing …

Understanding Disparities in Post Hoc Machine Learning Explanation

V Mhasawade, S Rahman, Z Haskell-Craig… - The 2024 ACM …, 2024‏ - dl.acm.org
Previous work has highlighted that existing post-hoc explanation methods exhibit disparities
in explanation fidelity (across “race” and “gender” as sensitive attributes), and while a large …

A Baseline Analysis of Reward Models' Ability To Accurately Analyze Foundation Models Under Distribution Shift

W LeVine, B Pikus, A Chen, S Hendryx - arxiv preprint arxiv:2311.14743, 2023‏ - arxiv.org
Foundation models, specifically Large Language Models (LLMs), have lately gained wide-
spread attention and adoption. Reinforcement Learning with Human Feedback (RLHF) …

Understanding subgroup performance differences of fair predictors using causal models

SR Pfohl, N Harris, C Nagpal, D Madras… - … 2023 Workshop on …, 2023‏ - openreview.net
A common evaluation paradigm compares the performance of a machine learning model
across subgroups to assess properties related to fairness. In this work, we argue that …

Not all distributional shifts are equal: Fine-grained robust conformal inference

J Ai, Z Ren - arxiv preprint arxiv:2402.13042, 2024‏ - arxiv.org
We introduce a fine-grained framework for uncertainty quantification of predictive models
under distributional shifts. This framework distinguishes the shift in covariate distributions …