Articles

Filter by type:

. Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data. arXiv, 2026.

PDF

. Position: agentic AI orchestration should be Bayes-consistent. ICML, 2026.

PDF

. REPO: Detoxifying LLMs via Representation Erasure-based Preference Optimization. CATS@ICML, 2026.

PDF OpenReview

. Less is More: Undertraining Experts Improves Model Upcycling. arXiv, 2025.

PDF

. Continual Learning in Vision-Language Models via Aligned Model Merging. arXiv, 2025.

PDF

. From Dormant to Deleted: Tamper-Resistant Unlearning Through Weight-Space Regularization. NeurIPS, 2025.

PDF OpenReview

. Leveraging Per-Instance Privacy for Machine Unlearning. ICML, 2025.

PDF PMLR OpenReview

. On Traceability in $\ell_p$ Stochastic Convex Optimization. NeurIPS, 2025.

PDF OpenReview

. The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws. ICLR, 2025.

PDF OpenReview

. Soup to go: mitigating forgetting during continual learning with model averaging. arXiv, 2025.

PDF

. Torque-Aware Momentum. arXiv, 2024.

PDF

. Improved Localized Machine Unlearning Through the Lens of Memorization. TMLR, 2024.

PDF OpenReview

. Unlearning in- vs. out-of-distribution data in LLMs under gradient-based methods. SafeGenAI@NeurIPS, 2024.

PDF OpenReview

. The Non-Local Model Merging Problem: Permutation Symmetries and Variance Collapse. arXiv, 2024.

PDF

. Mechanistic Unlearning: Robust Knowledge Unlearning and Editing via Mechanistic Localization. ICML, 2024.

PDF PMLR OpenReview

. Mixture of Experts in a Mixture of RL settings. RLC, 2024.

PDF OpenReview

. Are we making progress in unlearning? Findings from the first NeurIPS unlearning competition. arXiv, 2024.

PDF

. Data Selection for Transfer Unlearning. arXiv, 2024.

PDF

. SSFL: Discovering Sparse Unified Subnetworks at Initialization for Efficient Federated Learning. TMLR, 2024.

PDF OpenReview

. Simultaneous linear connectivity of neural networks modulo permutation. ECML PKDD, 2024.

PDF DOI

. Evaluating Interventional Reasoning Capabilities of Large Language Models. CaLM@NeurIPS, 2024.

PDF OpenReview

. Information Complexity of Stochastic Convex Optimization: Applications to Generalization, Memorization, and Tracing. ICML, 2024.

PDF PMLR OpenReview ICML Awards

. Mixtures of Experts Unlock Parameter Scaling for Deep RL. ICML, 2024.

PDF PMLR OpenReview

. Dataset Difficulty and the Role of Inductive Bias. arXiv, 2024.

PDF

. Leveraging Function Space Aggregation for Federated Learning at Scale. TMLR, 2023.

PDF OpenReview

. The Cost of Scaling Down Large Language Models: Reducing Model Size Affects Memory before In-context Learning. ICLR, 2023.

PDF OpenReview

. Identifying Spurious Biases Early in Training through the Lens of Simplicity Bias. AISTATS, 2023.

PDF PMLR OpenReview

. JaxPruner: A concise library for sparsity research. CPAL, 2023.

PDF Code OpenReview

. Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?. ICLR, 2023.

PDF

. Limitations of Information-Theoretic Generalization Bounds for Gradient Descent Methods in Stochastic Convex Optimization. ALT, 2023.

PDF

. The Effect of Data Dimensionality on Neural Network Prunability. NeurIPS ‘I Can’t Believe It’s Not Better’ Workshop: Understanding Deep Learning Through Empirical Falsification, 2022.

PDF

. Understanding Generalization via Leave-One-Out Conditional Mutual Information. ISIT, 2022.

PDF

. Pruning's Effect on Generalization Through the Lens of Training and Regularization. NeurIPS, 2022.

PDF

. Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks. NeurIPS, 2022.

PDF

. Towards a Unified Information-Theoretic Framework for Generalization. NeurIPS, 2021.

PDF

. Deep Learning on a Data Diet: Finding Important Examples Early in Training. NeurIPS, 2021.

PDF

. Information-Theoretic Generalization Bounds for Stochastic Gradient Descent. COLT, 2021.

PDF

. On the role of data in PAC-Bayes bounds. AISTATS, 2021.

PDF

. Pruning Neural Networks at Initialization: Why are We Missing the Mark?. ICLR, 2021.

PDF

. On the Information Complexity of Proper Learners for VC Classes in the Realizable Case. arXiv, 2020.

PDF

. Sharpened Generalization Bounds based on Conditional Mutual Information and an Application to Noisy, Iterative Algorithms. NeurIPS, 2020.

PDF

. Deep learning versus kernel learning: an empirical study of loss landscape geometry and the time evolution of the Neural Tangent Kernel. NeurIPS, 2020.

PDF

. In Search of Robust Measures of Generalization. NeurIPS, 2020.

PDF

. Enforcing Interpretability and its Statistical Impacts: Trade-offs between Accuracy and Interpretability. arXiv, 2020.

PDF

. Stochastic Neural Network with Kronecker Flow. AISTATS, 2020.

PDF

. RelatIF: Identifying Explanatory Training Examples via Relative Influence. AISTATS, 2020.

PDF

. Linear Mode Connectivity and the Lottery Ticket Hypothesis. ICML, 2020.

. In Defense of Uniform Convergence: Generalization via derandomization with an application to interpolating predictors. ICML, 2020.

PDF

. Information-Theoretic Generalization Bounds for SGLD via Data-Dependent Estimates. NeurIPS, 2019.

PDF

. Stabilizing the Lottery Ticket Hypothesis. arXiv, 2019.

PDF

. Data-dependent PAC-Bayes priors via differential privacy. NeurIPS, 2018.

PDF

. Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization properties of Entropy-SGD and data-dependent priors. ICML, 2018.

PDF

. Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data. UAI, 2017.

PDF

. A study of the effect of JPG compression on adversarial images. arXiv, 2016.

PDF

. Neural Network Matrix Factorization. arXiv, 2015.

PDF

. Requiem for the max rule?. Vision Research, 2015.

PDF

. Training generative neural networks via Maximum Mean Discrepancy optimization. UAI, 2015.

PDF