Sujay Sanghavi

Enhancing Language Model Agents using Diversity of Thoughts ICLR 2025 Geometric Median (GM) Matching for Robust k-Subset Selection from Noisy Data ICML 2025 Infilling Score: A Pretraining Data Detection Algorithm for Large Language Models ICLR 2025 InfoPO: On Mutual Information Maximization for Large Language Model Alignment NAACL 2025 Learning Mixtures of Experts with EM: A Mirror Descent Perspective ICML 2025 Retraining with Predicted Hard Labels Provably Increases Model Accuracy ICML 2025 Upweighting Easy Samples in Fine-Tuning Mitigates Forgetting ICML 2025 Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization NeurIPS 2024 DataComp-LM: In search of the next generation of training sets for language models NeurIPS 2024 Improving Computational Complexity in Statistical Models with Local Curvature Information ICML 2024 In-Context Learning with Transformers: Softmax Attention Adapts to Function Lipschitzness NeurIPS 2024 SVFT: Parameter-Efficient Fine-Tuning with Singular Vectors NeurIPS 2024 Time Weaver: A Conditional Time Series Generation Model ICML 2024 Understanding the Training Speedup from Sampling with Approximate Losses ICML 2024 Beyond Uniform Lipschitz Condition in Differentially Private Optimization ICML 2023 Finite-Time Logarithmic Bayes Regret Upper Bounds NeurIPS 2023 Latent Variable Representation for Reinforcement Learning ICLR 2023 Sample Efficiency of Data Augmentation Consistency Regularization AISTATS 2023 Understanding Self-Distillation in the Presence of Label Noise ICML 2023 Asymptotically-Optimal Gaussian Bandits with Side Observations ICML 2022 Faster non-convex federated learning via global and local momentum UAI 2022 Linear Bandit Algorithms with Sublinear Time Complexity ICML 2022 Minimax Regret for Cascading Bandits NeurIPS 2022 Robust Training in High Dimensions via Block Coordinate Geometric Median Descent AISTATS 2022 Towards Statistical and Computational Complexities of Polyak Step Size Gradient Descent AISTATS 2022 Toward Understanding Privileged Features Distillation in Learning-to-Rank NeurIPS 2022 Nearly Horizon-Free Offline Reinforcement Learning NeurIPS 2021 Choosing the Sample with Lowest Loss makes SGD Robust AISTATS 2020 Extreme Multi-label Classification from Aggregated Labels ICML 2020 Blocking Bandits NeurIPS 2019 Interaction Hard Thresholding: Consistent Sparse Quadratic Regression in Sub-quadratic Time and Space NeurIPS 2019 Iterative Least Trimmed Squares for Mixed Linear Regression NeurIPS 2019 Learning a Compressed Sensing Measurement Matrix via Gradient Unrolling ICML 2019 Learning Distributions Generated by One-Layer ReLU Networks NeurIPS 2019 Learning with Bad Training Data via Iterative Trimmed Loss Minimization ICML 2019 Sparse Logistic Regression Learns All Discrete Pairwise Graphical Models NeurIPS 2019 The Search Problem in Mixture Models JMLR 2018 Non-square matrix sensing without spurious local minima via the Burer-Monteiro approach AISTATS 2017 Dropping Convexity for Faster Semi-definite Optimization COLT 2016 Normalized Spectral Map Synchronization NeurIPS 2016 Single Pass PCA of Matrix Products NeurIPS 2016 Completing Any Low-rank Matrix, Provably JMLR 2015 Convergence Rates of Active Learning for Maximum Likelihood Estimation NeurIPS 2015 Preference Completion: Large-scale Collaborative Ranking from Pairwise Comparisons ICML 2015 Alternating Minimization for Mixed Linear Regression ICML 2014 Clustering Partially Observed Graphs via Convex Optimization JMLR 2014 Coherent Matrix Completion ICML 2014 Greedy Subspace Clustering NeurIPS 2014 Non-convex Robust PCA NeurIPS 2014 Phase Retrieval using Alternating Minimization NeurIPS 2013 Clustering Sparse Graphs NeurIPS 2012 On Learning Discrete Graphical Models using Group-Sparse Regularization AISTATS 2011 A Dirty Model for Multi-task Learning NeurIPS 2010 Robust PCA via Outlier Pursuit NeurIPS 2010 Linear programming analysis of loopy belief propagation for weighted matching NeurIPS 2007 Message Passing for Max-weight Independent Set NeurIPS 2007