Jennifer Dy

2026

1. 

   Inverted Detection and Control in Steering Vectors

   Max Torop, Aria Masoomi, and Jennifer Dy

   In Advances in Neural Information Processing Systems, 2026

2. 

   CD-RCM: Generalizable Continuous-Depth Novel View Synthesis for Reflectance Confocal Microscopy

   Tooba Imtiaz, Milind Rajadhyaksha, Kivanc Kose*, and Jennifer Dy*

   In Advances in Neural Information Processing Systems, 2026

3. 

   Feature Importance via Sets of Locally Performant Linear Models

   Fatemeh Tohidian, Davin Hill, Aria Masoomi, Peter J. Castaldi, and Jennifer Dy

   In Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, 2026

4. 

   ITSPACE: Monotone Gaussian Optimal Transport Updates

   Woojoo Na, and Jennifer Dy

   In International Conference on Machine Learning, ICML, 2026

2025

1. 

   DISCO: Disentangled Communication Steering for Large Language Models

   Max Torop, Aria Masoomi, Masih Eskandar, and Jennifer Dy

   In Advances in Neural Information Processing Systems, 2025

2. 

   OrdShap: Feature Position Importance for Sequential Black-Box Models

   Davin Hill, Brian L. Hill, Aria Masoomi, Vijay S. Nori, Robert E. Tillman, and Jennifer Dy

   2025

3. 

   LVT: Large-Scale Scene Reconstruction via Local View Transformers

   Tooba Imtiaz, Lucy Chai, Kathryn Heal, Xuan Luo, Jungyeon Park, Jennifer Dy, and John Flynn

   In Proceedings of the SIGGRAPH Asia 2025 Conference Papers, 2025

4. 

   Axiomatic Explainer Globalness via Optimal Transport

   Davin Hill, Josh Bone, Aria Masoomi, Max Torop, and Jennifer Dy

   In The 28th International Conference on Artificial Intelligence and Statistics, 2025

   Abs

   Explainability methods are often challenging to evaluate and compare. With a multitude of explainers available, practitioners must often compare and select explainers based on quantitative evaluation metrics. One particular differentiator between explainers is the diversity of explanations for a given dataset; i.e. whether all explanations are identical, unique and uniformly distributed, or somewhere between these two extremes. In this work, we define a complexity measure for explainers, globalness, which enables deeper understanding of the distribution of explanations produced by feature attribution and feature selection methods for a given dataset. We establish the axiomatic properties that any such measure should possess and prove that our proposed measure, Wasserstein Globalness, meets these criteria. We validate the utility of Wasserstein Globalness using image, tabular, and synthetic datasets, empirically showing that it both facilitates meaningful comparison between explainers and improves the selection process for explainability methods.

5. 

   STAR: Stability-Inducing Weight Perturbation for Continual Learning

   Masih Eskandar, Tooba Imtiaz, Davin Hill, Zifeng Wang, and Jennifer Dy

   In The Thirteenth International Conference on Learning Representations, 2025

2024

1. 

   Boundary-Aware Uncertainty for Feature Attribution Explainers

   Davin Hill, Aria Masoomi, Max Torop, Sandesh Ghimire, and Jennifer Dy

   In Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, 02–04 may 2024

   Abs PDF

   Post-hoc explanation methods have become a critical tool for understanding black-box classifiers in high-stakes applications. However, high-performing classifiers are often highly nonlinear and can exhibit complex behavior around the decision boundary, leading to brittle or misleading local explanations. Therefore there is an impending need to quantify the uncertainty of such explanation methods in order to understand when explanations are trustworthy. In this work we propose the Gaussian Process Explanation unCertainty (GPEC) framework, which generates a unified uncertainty estimate combining decision boundary-aware uncertainty with explanation function approximation uncertainty. We introduce a novel geodesic-based kernel, which captures the complexity of the target black-box decision boundary. We show theoretically that the proposed kernel similarity increases with decision boundary complexity. The proposed framework is highly flexible; it can be used with any black-box classifier and feature attribution method. Empirical results on multiple tabular and image datasets show that the GPEC uncertainty estimate improves understanding of explanations as compared to existing methods.

2. 

   Analyzing Explainer Robustness via Probabilistic Lipschitzness of Prediction Functions

   Zulqarnain Q Khan, Davin Hill, Aria Masoomi, Joshua T Bone, and Jennifer Dy

   In Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, 02–04 may 2024

   Abs PDF

   Machine learning methods have significantly improved in their predictive capabilities, but at the same time they are becoming more complex and less transparent. As a result, explainers are often relied on to provide interpretability to these black-box prediction models. As crucial diagnostics tools, it is important that these explainers themselves are robust. In this paper we focus on one particular aspect of robustness, namely that an explainer should give similar explanations for similar data inputs. We formalize this notion by introducing and defining explainer astuteness, analogous to astuteness of prediction functions. Our formalism allows us to connect explainer robustness to the predictor’s probabilistic Lipschitzness, which captures the probability of local smoothness of a function. We provide lower bound guarantees on the astuteness of a variety of explainers (e.g., SHAP, RISE, CXPlain) given the Lipschitzness of the prediction function. These theoretical results imply that locally smooth prediction functions lend themselves to locally robust explanations. We evaluate these results empirically on simulated as well as real datasets.

2023

1. 

   SmoothHess: ReLU Network Feature Interactions via Stein’s Lemma

   Max Torop, Aria Masoomi, Davin Hill, Kivanç Köse, Stratis Ioannidis, and Jennifer G. Dy

   In Advances in Neural Information Processing Systems, 02–04 may 2023

2. 

   DualHSIC: HSIC-Bottleneck and Alignment for Continual Learning

   Zifeng Wang, Zheng Zhan, Yifan Gong, Yucai Shao, Stratis Ioannidis, Yanzhi Wang, and Jennifer G. Dy

   In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, 02–04 may 2023

2022

1. 

   Deep Layer-wise Networks Have Closed-Form Weights

   Chieh Tzu Wu, Aria Masoomi, Arthur Gretton, and Jennifer G. Dy

   In International Conference on Artificial Intelligence and Statistics, AISTATS 2022, 28-30 March 2022, Virtual Event, 02–04 may 2022

2. 

   Learning to Prompt for Continual Learning

   Zifeng Wang, Zizhao Zhang, Chen-Yu Lee, Han Zhang, Ruoxi Sun, Xiaoqi Ren, Guolong Su, Vincent Perot, Jennifer G. Dy, and Tomas Pfister

   In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022, 02–04 may 2022

3. 

   Explanations of Black-Box Models based on Directional Feature Interactions

   Aria Masoomi, Davin Hill, Zhonghui Xu, Craig P. Hersh, Edwin K. Silverman, Peter J. Castaldi, Stratis Ioannidis, and Jennifer G. Dy

   In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022, 02–04 may 2022

4. 

   Explainable deep learning for insights in El Nino and river flows

   Yumin Liu, Kate Duffy, Jennifer G. Dy, and Auroop R. Ganguly

   CoRR, 02–04 may 2022

2021

1. 

   Segmentation of cellular patterns in confocal images of melanocytic lesions in vivo via a multiscale encoder-decoder network (MED-Net)

   Kivanç Köse, Alican Bozkurt, Christi Alessi-Fox, Melissa Gill, Caterina Longo, Giovanni Pellacani, Jennifer G. Dy, Dana H. Brooks, and Milind Rajadhyaksha

   Medical Image Anal., 02–04 may 2021

2. 

   Reliable Estimation of KL Divergence using a Discriminator in Reproducing Kernel Hilbert Space

   Sandesh Ghimire, Aria Masoomi, and Jennifer G. Dy

   In Advances in Neural Information Processing Systems, 02–04 may 2021

2020

1. 

   Open-World Class Discovery with Kernel Networks

   Zifeng Wang, Batool Salehi, Andrey Gritsenko, Kaushik R. Chowdhury, Stratis Ioannidis, and Jennifer G. Dy

   In 20th IEEE International Conference on Data Mining, ICDM 2020, Sorrento, Italy, November 17-20, 2020, 02–04 may 2020

2. 

   Instance-wise Feature Grouping

   Aria Masoomi, Chieh Wu, Tingting Zhao, Zifeng Wang, Peter J. Castaldi, and Jennifer G. Dy

   In Advances in Neural Information Processing Systems, 02–04 may 2020