Sewoong Oh

Concept-Aware Batch Sampling Improves Language-Image Pretraining CVPR 2026 Finite-Time Convergence Rates in Stochastic Stackelberg Games with Smooth Algorithmic Agents ICML 2025 PLeaS - Merging Models with Permutations and Least Squares CVPR 2025 Position: When Incentives Backfire, Data Stops Being Human ICML 2025 S4S: Solving for a Fast Diffusion Model Solver ICML 2025 Better Alignment with Instruction Back-and-Forth Translation EMNLP 2024 Can Public Large Language Models Help Private Cross-device Federated Learning? NAACL 2024 DataComp-LM: In search of the next generation of training sets for language models NeurIPS 2024 Data Mixture Inference Attack: BPE Tokenizers Reveal Training Data Compositions NeurIPS 2024 DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning ICML 2024 DPZero: Private Fine-Tuning of Language Models without Backpropagation ICML 2024 Improved Communication-Privacy Trade-offs in L_2 Mean Estimation under Streaming Differential Privacy ICML 2024 Insufficient Statistics Perturbation: Stable Estimators for Private Least Squares Extended Abstract COLT 2024 Multilingual Diversity Improves Vision-Language Representations NeurIPS 2024 One-shot Empirical Privacy Estimation for Federated Learning ICLR 2024 Privacy-Preserving Instructions for Aligning Large Language Models ICML 2024 Understanding the Gains from Repeated Self-Distillation NeurIPS 2024 CRISP: Curriculum based Sequential neural decoders for Polar code family ICML 2023 DataComp: In search of the next generation of multimodal datasets NeurIPS 2023 Few-shot Backdoor Attacks via Neural Tangent Kernels ICLR 2023 Improving multimodal datasets with image captioning NeurIPS 2023 Label Poisoning is All You Need NeurIPS 2023 Label Robust and Differentially Private Linear Regression: Computational and Statistical Efficiency NeurIPS 2023 Learning To Generate Image Embeddings With User-Level Differential Privacy CVPR 2023 MAUVE Scores for Generative Models: Theory and Practice JMLR 2023 On the Connection between Pre-training Data Diversity and Fine-tuning Robustness NeurIPS 2023 Private Federated Learning with Autotuned Compression ICML 2023 Private (Stochastic) Non-Convex Optimization Revisited: Second-Order Stationary Points and Excess Risks NeurIPS 2023 Unleashing the Power of Randomization in Auditing Differentially Private ML NeurIPS 2023 Why Is Public Pretraining Necessary for Private Model Training? ICML 2023 Bring Your Own Algorithm for Optimal Differentially Private Stochastic Minimax Optimization NeurIPS 2022 De novo mass spectrometry peptide sequencing with a transformer model ICML 2022 Differential privacy and robust statistics in high dimensions COLT 2022 DP-PCA: Statistically Optimal and Differentially Private PCA NeurIPS 2022 Eliminating Sharp Minima from SGD with Truncated Heavy-tailed Noise ICLR 2022 FedChain: Chained Algorithms for Near-optimal Communication Cost in Federated Learning ICLR 2022 Lifted Primal-Dual Method for Bilinearly Coupled Smooth Minimax Optimization AISTATS 2022 MAML and ANIL Provably Learn Representations ICML 2022 Quality Not Quantity: On the Interaction between Dataset Design and Robustness of CLIP NeurIPS 2022 Zonotope Domains for Lagrangian Neural Network Verification NeurIPS 2022 Divergence Frontiers for Generative Models: Sample Complexity, Quantization Effects, and Frontier Integrals NeurIPS 2021 Gradient Inversion with Generative Image Prior NeurIPS 2021 KO codes: inventing nonlinear encoding and decoding for reliable wireless communication via deep-learning ICML 2021 Robust and differentially private mean estimation NeurIPS 2021 SPECTRE: defending against backdoor attacks using robust statistics ICML 2021 Statistically and Computationally Efficient Linear Meta-representation Learning NeurIPS 2021 InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANs ICML 2020 Learning in Gated Neural Networks AISTATS 2020 Meta-learning for Mixed Linear Regression ICML 2020 Optimal transport mapping via input convex neural networks ICML 2020 Projection Efficient Subgradient Method and Optimal Nonsmooth Frank-Wolfe Method NeurIPS 2020 Robust Meta-learning for Mixed Linear Regression with Small Batches NeurIPS 2020 Breaking the gridlock in Mixture-of-Experts: Consistent and Efficient Algorithms ICML 2019 Efficient Algorithms for Smooth Minimax Optimization NeurIPS 2019 Iterative Bayesian Learning for Crowdsourced Regression AISTATS 2019 Learning One-hidden-layer Neural Networks under General Input Distributions AISTATS 2019 Minimax Optimal Estimation of Approximate Differential Privacy on Neighboring Databases NeurIPS 2019 Rate Distortion For Model Compression:From Theory To Practice ICML 2019 Spectrum Estimation from a Few Entries JMLR 2019 Turbo Autoencoder: Deep learning based channel codes for point-to-point communication channels NeurIPS 2019 Communication Algorithms via Deep Learning ICLR 2018 Deepcode: Feedback Codes via Deep Learning NeurIPS 2018 Generalized Rank-Breaking: Computational and Statistical Tradeoffs JMLR 2018 Learning from Comparisons and Choices JMLR 2018 PacGAN: The power of two samples in generative adversarial networks NeurIPS 2018 Robustness of conditional GANs to noisy labels NeurIPS 2018 Discovering Potential Correlations via Hypercontractivity NeurIPS 2017 Estimating Mutual Information for Discrete-Continuous Mixtures NeurIPS 2017 Matrix Norm Estimation from a Few Entries NeurIPS 2017 Optimal Sample Complexity of M-wise Data for Top-K Ranking NeurIPS 2017 Achieving budget-optimality with adaptive schemes in crowdsourcing NeurIPS 2016 Breaking the Bandwidth Barrier: Geometrical Adaptive Entropy Estimation NeurIPS 2016 Computational and Statistical Tradeoffs in Learning to Rank NeurIPS 2016 Conditional Dependence via Shannon Capacity: Axioms, Estimators and Applications ICML 2016 Data-driven Rank Breaking for Efficient Rank Aggregation ICML 2016 Data-driven Rank Breaking for Efficient Rank Aggregation JMLR 2016 Extremal Mechanisms for Local Differential Privacy JMLR 2016 Metadata-conscious anonymous messaging ICML 2016 Optimality of Belief Propagation for Crowdsourced Classification ICML 2016 Collaboratively Learning Preferences from Ordinal Data NeurIPS 2015 Secure Multi-party Differential Privacy NeurIPS 2015 The Composition Theorem for Differential Privacy ICML 2015 Extremal Mechanisms for Local Differential Privacy NeurIPS 2014 Learning Mixed Multinomial Logit Model from Ordinal Data NeurIPS 2014 Learning Mixtures of Discrete Product Distributions using Spectral Decompositions COLT 2014 Minimax-optimal Inference from Partial Rankings NeurIPS 2014 Provable Tensor Factorization with Missing Data NeurIPS 2014 Iterative ranking from pair-wise comparisons NeurIPS 2012 Iterative Learning for Reliable Crowdsourcing Systems NeurIPS 2011 Matrix Completion from Noisy Entries JMLR 2010 Matrix Completion from Noisy Entries NeurIPS 2009