Publications
($^*$ denotes alphabetical ordering or co-first author)
Preprints
R. Zhang, J. Wu, L. Lin, P. L. Bartlett. Minimax Optimal Convergence of Gradient Descent in Logistic Regression via Large and Adaptive Stepsizes. Preprint, 2025+. arXiv
L. Lin, S. Mei. A Statistical Theory of Contrastive Learning via Approximate Sufficient Statistics. Preprint, 2025+. arXiv
X. Zhao, W. Cai, T. Shi, D. Huang, L. Lin, S. Mei, D. Song. Improving LLM Safety Alignment with Dual-Objective Optimization. Preprint, 2025+. arXiv
K. Oko, L. Lin*, Y. Cai, S. Mei. A Statistical Theory of Contrastive Pre-training and Multimodal Generative AI. Preprint, 2025+. arXiv
L. Lin*, F. Su, W. Mou, P. Ding, M. J. Wainwright. When is it worthwhile to jackknife? Breaking the quadratic barrier for Z-estimators. Preprint, 2024+. arXiv
C. Fan, J. Liu, L. Lin*, J. Jia, R. Zhang, S. Mei, S. Liu. Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning. Preprint, 2024+. arXiv
M. Celentano, Z. Fan, L. Lin*, and S. Mei. Mean-field variational inference with the TAP free energy: Geometric and statistical properties in linear models. Preprint, 2023+. arXiv
T. Ahn, L. Lin*, S. Mei. Near-optimal multiple testing in Bayesian linear models with finite-sample FDR control. Preprint, 2022+. arXiv
Publications
L. Lin*, K. Khamaru, M. J. Wainwright. Semi-parametric inference based on adaptively collected data. To appear in Annals of Statistics, 2025. arXiv
L. Lin, J. Wu, S. M. Kakade, P. L. Bartlett, J. D. Lee. Scaling Laws in Linear Regression: Compute, Parameters, and Data. Advances in Neural Information Processing Systems (NeurIPS), 2024. arXiv
R. Zhang, L. Lin*, Y. Bai, S. Mei. Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning. The First Conference on Language Modeling (COLM), 2024. arXiv
L. Lin, T. Zrnic. Plug-in Performative Optimization. International Conference on Machine Learning (ICML), 2024. arXiv
L. Lin, Y. Bai, S. Mei. Transformers as Decision Makers: Provable In-Context Reinforcement Learning via Supervised Pretraining. International Conference on Learning Representations (ICLR), 2024. arXiv
L. Lin, M. Ying, S. Ghosh, K. Khamaru, and C. Zhang. Statistical Limits of Adaptive Linear Models: Low-Dimensional Estimation and Inference. Advances in Neural Information Processing Systems (NeurIPS), 2023. arXiv
L. Lin, E. Dobriban. What causes the test error? going beyond bias-variance via anova. The Journal of Machine Learning Research, 2021. journal arXiv