Conference papers

• ABC-Align: Prediction-Powered Alignment with Adaptive Bias Control
  Eric Frankel, Banghua Zhu, Sewoong Oh, Lillian J. Ratliff
  working paper

  • code for post-training LLMs with ABC-Align

• MIRAGE: Protecting against Malicious Image Editing via False Moderation
  Anshul Nasery, Ramnath Kumar, Cho-Jui Hsieh, Sewoong Oh
  working paper

• Foundation model for mass spectrometry proteomics
  Justin Sanders, Melih Yilmaz, Jacob H. Russell, Wout Bittremieux, William E. Fondrie, Nicholas M. Riley, Sewoong Oh, William Stafford Noble
  working paper

  • Workshop on Multi-modal Foundation Models and Large Language Models for Life Sciences at ICML 2025
    

• Correct Answers from Sound Reasoning: Verifiable Process Supervision for Language Models
  Kyuyoung Kim, Kevin Wang, Yunfei Xie, Peiyang Xu, Peiyao Sheng, Chen Wei, Zhangyang Wang, Jinwoo Shin, Pramod Viswanath, Sewoong Oh
  COLM 2026
  

• Zeroth-Order Optimization at the Edge of Stability
  Minhak Song, Liang Zhang, Bingcong Li, Niao He, Michael Muehlebach, Sewoong Oh
  ICML 2026
  

  • Oral presentation at Sci4DL workshop at ICLR 2026
    

  • ICML poster
    

• Sampling from Your Language Model One Byte at a Time
  Jonathan Hayase, Alisa Liu, Noah A. Smith, Sewoong Oh
  ICML 2026
  

  • TokShop tokenization workshop at ICML 2025
    

  • code for Byte-sampling from your LLM

• Spurious Rewards: Rethinking Training Signals in RLVR
  Rulin Shao, Shuyue Stella Li, Rui Xin, Scott Geng, Yiping Wang, Sewoong Oh, Simon Shaolei Du, Nathan Lambert, Sewon Min, Ranjay Krishna, Yulia Tsvetkov, Hannaneh Hajishirzi, Pang Wei Koh, Luke Zettlemoyer
  ICML 2026

• Anchored Decoding: Provably Reducing Copyright Risk for Any Language Model
  Jacqueline He, Jonathan Hayase, Wen-tau Yih, Sewoong Oh, Luke Zettlemoyer, Pang Wei Koh
  ICML 2026
  

  • an Honorable Mention for the best paper selection at DATA-FM workshop at ICLR’26
    

• Privasis: Synthesizing the Largest “Public” Private Dataset from Scratch
  Hyunwoo Kim, Niloofar Mireshghallah, Michael Duan, Rui Xin, Shuyue Stella Li, Jaehun Jung, David Acuna, Qi Pang, Hanshen Xiao, G. Edward Suh, Sewoong Oh, Yulia Tsvetkov, Pang Wei Koh, Yejin Choi
  ICML 2026

• Concept-Aware Batch Sampling Improves Language-Image Pretraining
  Adhiraj Ghosh, Vishaal Udandarao, Thao Nguyen, Matteo Farina, Mehdi Cherti, Jenia Jitsev, Sewoong Oh, Elisa Ricci, Ludwig Schmidt, Matthias Bethge
  CVPR 2026

• Open Deep Search: Democratizing Search with Open-source Reasoning Agents
  Salaheddin Alzubi, Creston Brooks, Purva Chiniya, Edoardo Contente, Chiara von Gerlach, Lucas Irwin, Yihan Jiang, Arda Kaz, Windsor Nguyen, Sewoong Oh, Himanshu Tyagi, Pramod Viswanath
  technical report

• OpenThoughts: Data Recipes for Reasoning Models
  Etash Guha, Ryan Marten, Sedrick Keh, Negin Raoof, Georgios Smyrnis, Hritik Bansal, Marianna Nezhurina, Jean Mercat, Trung Vu, Zayne Sprague, Ashima Suvarna, Benjamin Feuer, Liangyu Chen, Zaid Khan, Eric Frankel, Sachin Grover, Caroline Choi, Niklas Muennighoff, Shiye Su, Wanjia Zhao, John Yang, Shreyas Pimpalgaonkar, Kartik Sharma, Charlie Cheng-Jie Ji, Yichuan Deng, Sarah Pratt, Vivek Ramanujan, Jon Saad-Falcon, Jeffrey Li, Achal Dave, Alon Albalak, Kushal Arora, Blake Wulfe, Chinmay Hegde, Greg Durrett, Sewoong Oh, Mohit Bansal, Saadia Gabriel, Aditya Grover, Kai-Wei Chang, Vaishaal Shankar, Aaron Gokaslan, Mike A. Merrill, Tatsunori Hashimoto, Yejin Choi, Jenia Jitsev, Reinhard Heckel, Maheswaran Sathiamoorthy, Alexandros G. Dimakis, Ludwig Schmidt
  ICLR 2026

• Are Robust LLM Fingerprints Adversarially Robust?
  Anshul Nasery, Edoardo Contente, Alkin Kaz, Pramod Viswanath, Sewoong Oh
  SatML 2026

• A False Sense of Privacy: Evaluating Textual Data Sanitization Beyond Surface-level Privacy Leakage
  Rui Xin, Niloofar Mireshghallah, Shuyue Stella Li, Michael Duan, Hyunwoo Kim, Yejin Choi, Yulia Tsvetkov, Sewoong Oh, Pang Wei Koh
  SatML 2026

• Scalable Fingerprinting of Large Language Models
  Anshul Nasery, Jonathan Hayase, Creston Brooks, Peiyao Sheng, Himanshu Tyagi, Pramod Viswanath, Sewoong Oh
  NeurIPS 2025, (Spolight presentation)
  

  • present at ICLR 2025 workshop on GenAI Watermarking
    

• Understanding the Gain from Data Filtering in Multimodal Contrastive Learning
  Divyansh Pareek, Sewoong Oh, Simon Shaolei Du
  NeurIPS 2025

• Zeroth-Order Optimization Finds Flat Minima
  Liang Zhang, Bingcong Li, Kiran Koshy Thekumparampil, Sewoong Oh, Michael Muehlebach, Niao He
  NeurIPS 2025

• Recycling the Web: A Method to Enhance Pre-training Data Quality and Quantity for Language Models
  Thao Nguyen, Yang Li, Olga Golovneva, Luke Zettlemoyer, Sewoong Oh, Ludwig Schmidt, Xian Li
  COLM 2025

  • DataWorld workshop at ICML 2025
    

• SuperBPE: Space Travel for Language Models
  Alisa Liu, Jonathan Hayase, Valentin Hofmann, Sewoong Oh, Noah A. Smith, Yejin Choi
  COLM 2025

  • project website

  • code for superBPE tokenizer training

  • OLMo-2 8B and 11B models trained with SuperBPE

  • presented by Sewoong at AIMACCS workshop (slides)
    

  • presented by Sewoong at ICLR 2025 Synthetic Data × Data Access Problem workshop (slides)
    

  • presented by Alisa and Jon at Microsoft (slides)
    

  • TokShop tokenization workshop at ICML 2025

• Economics of Sourcing Human Data
  Sebastin Santy, Prasanta Bhattacharya, Manoel Horta Ribeiro, Kelsey Allen, Sewoong Oh
  ICML 2025, position paper track

• Convergence Rates in Stochastic Stackelberg Games with Smooth Algorithmic Agents
  Eric Frankel, Kshitij Kulkarni, Dmitriy Drusvyatskiy, Sewoong Oh, Lillian J. Ratliff
  ICML 2025

• S4S: Solving for a Diffusion Model Solver
  Eric Frankel, Sitan Chen, Jerry Li, Pang Wei Koh, Lillian J. Ratliff, Sewoong Oh
  ICML 2025

• PLeaS–Merging Models with Permutations and Least Squares
  Anshul Nasery, Jonathan Hayase, Pang Wei Koh, Sewoong Oh
  CVPR 2025

• Data Mixture Inference: What do BPE Tokenizers Reveal about their Training Data?
  Jonathan Hayase, Alisa Liu, Yejin Choi, Sewoong Oh, Noah A. Smith
  NeurIPS 2024
  

  • presented at ICML 2024 workshop on Data-centric Machine Learning Research (DMLR)

• Multilingual Diversity Improves Vision-Language Representations
  Thao Nguyen, Matthew Wallingford, Sebastin Santy, Wei-Chiu Ma, Sewoong Oh, Ludwig Schmidt, Pang Wei Koh, Ranjay Krishna
  NeurIPS 2024

  • present by Sewoong at ICLR 2025 Synthetic Data × Data Access Problem workshop (slides)
    

• DataComp-LM: In search of the next generation of training sets for language models DataComp-LM: In search of the next generation of training sets for language models
  Jeffrey Li, Alex Fang, Georgios Smyrnis, Maor Ivgi, Matt Jordan, Samir Gadre, Hritik Bansal, Etash Guha, Sedrick Keh, Kushal Arora, Saurabh Garg, Rui Xin, Niklas Muenninghoff, Reinhard Heckel, Jean Mercat, Mayee Chen, Suchin Gururangan, Mitchell Wortsman, Alon Albalak, Yonatan Bitton, Marianna Nezhurina, Amro Abbas, Cheng-Yu Hsieh, Dhruba Ghosh, Josh Gardner, Maciej Kilian, Hanlin Zhang, Rulin Shao, Sarah Pratt, Sunny Sanyal, Gabriel Ilharco, Giannis Daras, Kalyani Marathe, Aaron Gokaslan, Jieyu Zhang, Khyathi Chandu, Thao Nguyen, Igor Vasiljevic, Sham Kakade, Shuran Song, Sujay Sanghavi, Fartash Faghri, Sewoong Oh, Luke Zettlemoyer, Kyle Lo, Alaaeldin El-Nouby, Hadi Pouransari, Alexander Toshev, Stephanie Wang, Dirk Groeneveld, Luca Soldani, Pang Wei Koh, Jenia Jitsev, Thomas Kollar, Alexandros G Dimakis, Yair Carmon, Achal Dave, Ludwig Schmidt, Vaishaal Shankar
  NeurIPS 2024, Datasets and Benchmarks Track

  • Check out the competition at datacomp.ai

• Understanding the Gains from Repeated Self-Distillation
  Divyansh Pareek, Simon S. Du, Sewoong Oh
  NeurIPS 2024
  

  • presented at ICML 2024 workshop on Data-centric Machine Learning Research (DMLR)

• Better Alignment with Instruction Back-and-Forth Translation
  Thao Nguyen, Jeffrey Li, Sewoong Oh, Ludwig Schmidt, Jason Weston, Luke Zettlemoyer, Xian Li
  EMNLP 2024

• Air Gap: Protecting Privacy-Conscious Conversational Agents
  Eugene Bagdasaryan, Ren Yi, Sahra Ghalebikesabi, Peter Kairouz, Marco Gruteser, Sewoong Oh, Borja Balle, Daniel Ramage
  CCS 2024

• Insufficient Statistics Perturbation: Stable Estimators for Private Least Squares
  Gavin Brown, Jonathan Hayase, Samuel Hopkins, Weihao Kong, Xiyang Liu, Sewoong Oh, Juan C. Perdomo, Adam Smith
  COLT, 2024
  

  • presented at FORC 2024, Symposium on the Fooundations of Responsible Computing

  • presented at TPDP 2024, Theory and Practice of Differential Privacy

  • slides presented at Mathematical Foundations of Machine Learning, Monte Verita Congress Center, Switzerland, 2025

• DPZero: Private Fine-Tuning of Language Models without Backpropagation
  Liang Zhang, Bingcong Li, Kiran Koshy Thekumparampil, Sewoong Oh, Niao He
  ICML, 2024
  

  • presented at TPDP 2024, Theory and Practice of Differential Privacy

  • slides from my talk at CSL student conference at UIUC are available here

  • slides from my talk at IMS International Conference on Statistics and Data Science (ICSDS) 2023 are available here

  • a workshop version presented at NeurIPS 2023 workshop on Federated Learning in the Age of Foundation Models

    • DPZero: Dimension-Independent and Differentially Private Zeroth-Order Optimization, Liang Zhang, Kiran Koshy Thekumparampil, Sewoong Oh, Niao He
      

• Improved Communication-Privacy Trade-offs in Mean Estimation under Streaming Differential Privacy
  Wei-Ning Chen, Berivan Isik, Peter Kairouz, Albert No, Sewoong Oh, Zheng Xu
  ICML, 2024
  

• Privacy-Preserving Instructions for Aligning Large Language Models
  Da Yu, Peter Kairouz, Sewoong Oh, Zheng Xu
  ICML, 2024
  

• DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning
  S Ashwin Hebbar, Sravan Kumar Ankireddy, Hyeji Kim, Sewoong Oh, Pramod Viswanath
  ICML, 2024
  

• Uncertainty Quantification with User-level Differential Privacy
  Abhradeep Guha Thakurta, Dj Dvijotham, Georgie Evans, Peter Kairouz, Ryan McKenna, Sewoong Oh
  working paper
  

  • present at Theory and Practice of Differential Privacy 2023

• Profit: Benchmarking Personalization and Robustness Trade-off in Federated Prompt Tuning
  Collins, Liam, Shanshan Wu, Sewoong Oh, and Khe Chai Sim
  

  • present at International Workshop on Federated Learning in the Age of Foundation Models in Conjunction with NeurIPS 2023

• Can Public Large Language Models Help Private Cross-device Federated Learning?
  Boxin Wang, Yibo Jacky Zhang, Yuan Cao, Bo Li, H Brendan McMahan, Sewoong Oh, Zheng Xu, Manzil Zaheer
  NAACL, 2024
  

  • presented at ICML 2023 workshop on Efficient Systems for Foundation Models

  • presented at ICML 2023 workshop on Challenges of Deploying Generative AI

• One-shot Empirical Privacy Estimation for Federated Learning
  Andrew, Galen, Peter Kairouz, Sewoong Oh, Alina Oprea, H. Brendan McMahan, and Vinith Suriyakumar
  ICLR 2024, (Oral presentation)
  

  • International Workshop on Federated Learning in the Age of Foundation Models in Conjunction with NeurIPS 2023

• Advancing Differential Privacy: Where We Are Now and Future Directions for Real-World Deployment
  Rachel Cummings, Damien Desfontaines, David Evans, Roxana Geambasu, Matthew Jagielski, Yangsibo Huang, Peter Kairouz, Gautam Kamath, Sewoong Oh, Olga Ohrimenko, Nicolas Papernot, Ryan Rogers, Milan Shen, Shuang Song, Weijie Su, Andreas Terzis, Abhradeep Thakurta, Sergei Vassilvitskii, Yu-Xiang Wang, Li Xiong, Sergey Yekhanin, Da Yu, Huanyu Zhang, Wanrong Zhang
  Harvard Data Science Review, 6(1), January 2024
  

  • In July 2022, we hosted a workshop titled “Differential Privacy (DP): Challenges towards the Next Frontier” with experts from industry, academia, and the public sector to discuss and find solutions to the challenges of differential privacy. This is a report from that workshop.

• Label Poisoning is All You Need
  Rishi D. Jha, Jonathan Hayase, Sewoong Oh
  NeurIPS, 2023
  

  • presented at NeurIPS 2023 workshop on Federated Learning in the Age of Foundation Models

• Unleashing the power of randomization in auditing differentially private ML
  Krishna Pillutla, Galen Andrew, Peter Kairouz, Brendan McMahan, Alina Oprea, Sewoong Oh
  NeurIPS 2023
  

  • presented at Boston Privacy Day, Google, and MSR

  • presented at ICML 2023 Federated Learning workshop

  • Presented at CRYPTO 2023 Privacy Preserving Machine Learning (PPML) workshop

• Near Optimal Private and Robust Linear Regression
  Xiyang Liu, Prateek Jain, Weihao Kong, Sewoong Oh, Arun Sai Suggala
  NeurIPS 2023

• Private (Stochastic) Non-Convex Optimization Revisited: Second-Order Stationary Points and Excess Risks
  Arun Ganesh, Daogao Liu, Sewoong Oh, Abhradeep Thakurta
  NeurIPS 2023 (Spotlight presentation)

• On the Connection between Pre-training Data Diversity and Fine-tuning Robustness
  Vivek Ramanujan, Thao Nguyen, Sewoong Oh, Ludwig Schmidt, Ali Farhadi
  NeurIPS 2023 (Spotlight presentation)

• Improving multimodal datasets with image captioning
  Thao Nguyen, Samir Yitzhak Gadre, Gabriel Ilharco, Sewoong Oh, Ludwig Schmidt
  NeurIPS 2023, Datasets and Benchmarks Track
  

  • presented at ICML 2023 workshop on Data-centric Machine Learning Research (DMLR)

• DataComp: In search of the next generation of multimodal datasets
  Samir Yitzhak Gadre, Gabriel Ilharco, Alex Fang, Jonathan Hayase, Georgios Smyrnis, Thao Nguyen, Ryan Marten, Mitchell Wortsman, Dhruba Ghosh, Jieyu Zhang, Eyal Orgad, Rahim Entezari, Giannis Daras, Sarah Pratt, Vivek Ramanujan, Yonatan Bitton, Kalyani Marathe, Stephen Mussmann, Richard Vencu, Mehdi Cherti, Ranjay Krishna, Pang Wei Koh, Olga Saukh, Alexander Ratner, Shuran Song, Hannaneh Hajishirzi, Ali Farhadi, Romain Beaumont, Sewoong Oh, Alex Dimakis, Jenia Jitsev, Yair Carmon, Vaishaal Shankar, Ludwig Schmidt
  NeurIPS 2023, Datasets and Benchmarks Track (Oral presentation)
  

  • check out our benchmark competition at datacomp.ai

• Private federated learning with autotuned compression
  Enayat Ullah, Christopher Choquette, Peter Kairouz, Sewoong Oh
  ICML 2023,

• Why Is Public Pretraining Necessary for Private Model Training?
  Arun Ganesh, Mahdi Haghifam, Milad Nasr, Sewoong Oh, Thomas Steinke, Om Thakkar, Abhradeep Thakurta, Lun Wang
  ICML 2023,

• CRISP: Curriculum based Sequential Neural Decoders for Polar Code Family
  S Ashwin Hebbar, Viraj Nadkarni, Ashok Vardhan Makkuva, Suma Bhat, Sewoong Oh, Pramod Viswanath
  ICML 2023,

• Learning to Generate Image Embeddings with User-level Differential Privacy
  Zheng Xu, Maxwell Collins, Yuxiao Wang, Liviu Panait, Sewoong Oh, Sean Augenstein, Ting Liu, Florian Schroff, H. Brendan McMahan
  CVPR 2023,

• Few-shot Backdoor Attacks via Neural Tangent Kernels
  Jonathan Hayase, Sewoong Oh
  ICLR, 2023

  • poster presented at NeurIPS 2022 workshop on Trustworthy and Socially Responsible Machine Learning (TSRML)

• Zonotope Domains for Lagrangian Neural Network Verification
  Matt Jordan, Jonathan Hayase, Alexandros G Dimakis, Sewoong Oh
  NeurIPS 2022,

• Quality Not Quantity: On the Interaction between Dataset Design and Robustness of CLIP
  Thao Nguyen, Gabriel Ilharco, Mitchell Wortsman, Sewoong Oh, Ludwig Schmidt
  NeurIPS 2022, (Oral presentation)

  • presented at ICML 2022 workshop on Benchmarking Data for Data-Centric AI (DataPerf)

• Bring Your Own Algorithm for Optimal Differentially Private Stochastic Minimax Optimization
  Liang Zhang, Kiran Koshy Thekumparampil, Sewoong Oh, Niao He
  NeurIPS 2022,

• DP-PCA: Statistically Optimal and Differentially Private PCA
  Xiyang Liu, Weihao Kong, Prateek Jain, Sewoong Oh
  NeurIPS 2022,

• MAML and ANIL Provably Learn Representations
  Liam Collins, Aryan Mokhtari, Sewoong Oh, Sanjay Shakkottai
  ICML, 2022

  • video of a talk on Dec 2022 at C3 Digital Transformation Institute is available here

  • slides from my talk at C3 Digital Transformation Institute is available here

  • 5 minutes presentation by Liam Collins at ICML is available here

• De novo mass spectrometry peptide sequencing with a transformer model
  Melih Yilmaz, William E. Fondrie, Wout Bittremieux, Sewoong Oh, William Stafford Noble
  ICML, 2022

  • 5 minutes presentation by Melih Yilmaz at ICML is available here

• Differential privacy and robust statistics in high dimensions
  Xiyang Liu, Weihao Kong, Sewoong Oh
  COLT, 2022

  • Presented at the third AAAI Workshop on Privacy-Preserving Artificial Intelligence (PPAI-22), the recording of a 12 minute presentation is available here

  • video of a talk on Nov 2021 at SNAPP seminar series is available here

  • slides from my talk at SNAPP seminar is available here

  • slides from my talk at Google is available here

• Lifted Primal-Dual Method for Bilinearly Coupled Smooth Minimax Optimization
  Kiran Koshy Thekumparampil, Niao He, Sewoong Oh
  AISTATS, 2022 (Oral presentation)

  • video of a presention on Feb 2022 at Simons institute by Niao He is here

• Eliminating Sharp Minima from SGD with Truncated Heavy-tailed Noise
  Xingyu Wang, Sewoong Oh, Chang-Han Rhee
  ICLR, 2022

  • presented at DeepMath 2021

  • video of a talk Xingyu gave at DeepMath is available here

  • slides from Xingyu's talk is available here

  • video of a talk on Feb 2022 at SNAPP seminar series by Chang-han is here

• FedChain: Chained Algorithms for Near-Optimal Communication Cost in Federated Learning
  Charlie Hou, Kiran K. Thekumparampil, Giulia Fanti, Sewoong Oh
  ICLR, 2022

  • presented at the ICML 2021 Workshop on Federated Learning for User Privacy and Data Confidentiality (ICML-FL 2021)

• Robust and Differentially Private Mean Estimation
  Xiyang Liu, Weihao Kong, Sham Kakade, Sewoong Oh
  NeurIPS 2021,

  • presented at the ICML 2021 Workshop on Federated Learning for User Privacy and Data Confidentiality (ICML-FL 2021)

  • presented at the CCS 2021 workshop Privacy Preserving Machine Learning (PPML’21)

  • video of a talk on Oct 2021 at Simons Institute is available here

  • slides from my talk is available here

  • code is available here

• Sample Efficient Linear Meta-Learning by Alternating Minimization
  Kiran Koshy Thekumparampil, Prateek Jain, Praneeth Netrapalli, Sewoong Oh
  NeurIPS, 2021

• Divergence Frontiers for Generative Models: Sample Complexity, Quantization Level, and Frontier Integral
  Lang Liu, Krishna Pillutla, Sean Welleck, Sewoong Oh, Yejin Choi, Zaid Harchaoui
  NeurIPS, 2021

• Gradient Inversion with Generative Image Prior
  Jaechang Kim, Jinwoo Jeon, Kangwook Lee, Sewoong Oh, and Jungseul Ok,
  NeurIPS, 2021

  • presented at the ICML 2021 Workshop on Federated Learning for User Privacy and Data Confidentiality (ICML-FL 2021)

• SPECTRE: Defending against backdoor attacks using robust covariance estimation
  Jonathan Hayase, Weihao Kong, Raghav Somani, Sewoong Oh
  ICML, 2021
  

  • you can find the code for SPECTRE here.

  • Backdoor-Suite: a user-friendly repository for backdoor attacks and defenses is available here

• KO codes: Inventing Nonlinear Encoding and Decoding for Reliable WirelessCommunication via Deep-Learning
  Ashok Vardhan Makkuva, Xiyang Liu, Mohammad Vahid Jamali, Hessam Mahdavifar, Sewoong Oh, Pramod Viswanath
  ICML, 2021
  

  • slides from a talk by Ashok is available here: 5 min and 1 hour

• Reed-Muller Subcodes: Machine Learning-Aided Design of Efficient Soft Recursive Decoding
  Mohammad Vahid Jamali, Xiyang Liu, Ashok Vardhan Makkuva, Hessam Mahdavifar, Sewoong Oh, Pramod Viswanath
  ISIT, 2021

• Projection Efficient Subgradient Method and Optimal Nonsmooth Frank-Wolfe Method
  Kiran Koshy Thekumparampil, Prateek Jain, Praneeth Netrapalli, Sewoong Oh
  NeurIPS, 2020 (Spotlight presentation)

• Robust Meta-learning for Mixed Linear Regression with Small Batches
  Weihao Kong, Raghav Somani, Sham Kakade, Sewoong Oh
  NeurIPS, 2020

  • you can find the code for Robust PCA here.

• Optimal transport mapping via input convex neural networks
  Ashok Vardhan Makkuva, Amirhossein Taghvaei, Sewoong Oh, Jason D. Lee
  ICML, 2020

• Meta-learning for mixed linear regression
  Weihao Kong, Raghav Somani, Zhao Song, Sham Kakade, Sewoong Oh
  ICML, 2020

• InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANs
  Zinan Lin, Kiran Koshy Thekumparampil, Giulia Fanti, Sewoong Oh
  ICML, 2020

• Learning in Gated Neural Networks
  Ashok Vardhan Makkuva, Sewoong Oh, Sreeram Kannan, and Pramod Viswanath
  AISTATS, 2020

• Turbo Autoencoder: Deep learning based channel code for point-to-point communication channels
  Yihan Jiang, Hyeji Kim, Himanshu Asnani, Sreeram Kannan, Sewoong Oh, Pramod Viswanath
  NeurIPS, 2019

• Efficient Algorithms for Smooth Minimax Optimization
  Kiran Koshy Thekumparampil, Prateek Jain, Praneeth Netrapalli, Sewoong Oh
  NeurIPS, 2019
  

  • you can find the code for DIAG here

• Minimax Rates of Estimating Approximate Differential Privacy
  Xiyang Liu, Sewoong Oh
  NeurIPS, 2019

• Rate Distortion For Model Compression:From Theory To Practice
  Weihao Gao, Yu-Han Liu, Chong Wang, Sewoong Oh
  ICML, 2019

• Breaking the gridlock in Mixture-of-Experts: Consistent and Efficient Algorithms
  Ashok Vardhan Makkuva, Sewoong Oh, Sreeram Kannan, and Pramod Viswanath
  ICML, 2019
  

  • Slides from Ashok's presentation

• DeepTurbo: Deep Turbo Decoder
  Yihan Jiang, Hyeji Kim, Himanshu Asnani, Sreeram Kannan, Sewoong Oh, and Pramod Viswanath
  2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), 2019

• Barracuda: The Power of l-polling in Proof-of-Stake Blockchains
  G Fanti, J Jiao, A Makkuva, S Oh, R Rana, and P Viswanath
  ACM MobiHoc, 2019, (Best paper award)

• LEARN Codes: Inventing Low-latency Codes via Recurrent Neural Networks
  Yihan Jiang, Hyeji Kim, Himanshu Asnani, Sreeram Kannan, Sewoong Oh, and Pramod Viswanath
  IEEE International Conference on Communications (ICC), 2019

• Learning One-hidden-layer Neural Networks under General Input Distributions
  Weihao Gao, Ashok Vardhan Makkuva, Sewoong Oh, and Pramod Viswanath
  AISTATS, 2019

• Iterative Bayesian Learning for Crowdsourced Regression
  Jungseul Ok, Yunhun Jang, Sewoong Oh, Jinwoo Shin, Yung Yi
  AISTATS, 2019, [ code ]

• Compounding of Wealth in Proof-of-Stake Cryptocurrencies
  Giulia Fanti, Leonid Kogan, Sewoong Oh, Kathleen Ruan, Pramod Viswanath, and Gerui Wang
  Financial Cryptography and Data Security, 2019

• Robustness of conditional GANs to noisy labels
  Kiran Thekumparampil, Ashish Khetan, Zinan Lin, Sewoong Oh
  NIPS, 2018 (Spotlight presentation), [ code ]

• Deepcode: Feedback Codes via Deep Learning
  Hyeji Kim, Yihan Jiang, Sreeram Kannan, Sewoong Oh, Pramod Viswanath
  NIPS, 2018, [ code by Hyeji Kim ], [ code by Yihan Jiang ]

• PacGAN: The power of two samples in generative adversarial networks
  Zinan Lin, Ashish Khetan, Giulia Fanti, Sewoong Oh
  NIPS, 2018, [ code ], [ project page]

• Communication Algorithms via Deep Learning
  H. Kim, Y. Jiang, R. B. Rana, S. Kannan, S. Oh, and P. Viswanath
  ICLR, 2018

• Estimating Mutual Information for Discrete-Continuous Mixtures
  Weihao Gao, Sreeram Kannan, Sewoong Oh, and Pramod Viswanath
  NIPS, 2017 (Spotlight presentation) [ code ]

• Discovering Potential Correlations via Hypercontractivity
  Hyeji Kim, Weihao Gao, Sreeram Kannan, Sewoong Oh, and Pramod Viswanath
  NIPS, 2017 [ code ]

• Matrix Norm Estimation from a Few Entries
  Ashish Khetan, Sewoong Oh
  NIPS, 2017, (Spotlight presentation) [ code ]

• Top-K Ranking from Pairwise Comparisons: When Spectral Ranking is Optimal
  Minje Jang, Sunghyun Kim, Changho Suh, Sewoong Oh
  NIPS, 2017

• Density Functional Estimators with k-Nearest Neighbor Bandwidths
  Weihao Gao, Sewoong Oh, Pramod Viswanath
  ISIT, 2017

• Demystifying Fixed k-Nearest Neighbor Information Estimators
  Weihao Gao, Sewoong Oh, Pramod Viswanath
  ISIT, 2017, [ code ]

• Achieving budget-optimality with adaptive schemes in crowdsourcing
  Ashish Khetan, Sewoong Oh
  NIPS, 2016

• Computational and Statistical Tradeoffs in Learning to Rank
  Ashish Khetan, Sewoong Oh
  NIPS, 2016

• Breaking the Bandwidth Barrier: Geometrical Adaptive Entropy Estimation
  Weihao Gao, Sewoong Oh, Pramod Viswanath
  NIPS, 2016, [ code ]

• Conditional Dependence via Shannon Capacity: Axioms, Estimators and Applications
  Weihao Gao, Sreeram Kannan, Sewoong Oh, Pramod Viswanath
  ICML, 2016, [ code ]

• Data-driven Rank Breaking for Efficient Rank Aggregation
  Ashish Khetan, Sewoong Oh
  ICML, 2016

• Optimality of Belief Propagation for Crowdsourced Classification
  Jungseul Ok, Sewoong Oh, Jinwoo Shin, Yung Yi
  ICML, 2016

• Metadata-conscious Anonymous Messaging
  Giulia Fanti, Peter Kairouz, Sewoong Oh, Kannan Ramchandran and Pramod Viswanath
  ICML 2016

• Rumor Source Obfuscation on Irregular Trees
  Giulia Fanti, Peter Kairouz, Sewoong Oh, Kannan Ramchandran and Pramod Viswanath
  SIGMETRICS 2016

• Collaboratively Learning Preferences from Ordinal Data
  Sewoong Oh, Kiran K. Thekumparampil, and Jiaming Xu
  NIPS 2015

• Detecting Sponsored Recommendations Subhashini Krishnasamy, Rajat Sen, Sewoong Oh, and Sanjay Shakkottai
  SIGMETRICS (short paper) 2015

• Spy vs. Spy: Rumor Source Obfuscation
  Giulia Fanti, Peter Kairouz, Sewoong Oh, and Pramod Viswanath
  SIGMETRICS 2015 (Best paper award)

• Secure Multi-party Differential Privacy
  Peter Kairouz, Sewoong Oh, and Pramod Viswanath
  NIPS 2015

• Extremal Mechanisms for Local Differential Privacy
  Peter Kairouz, Sewoong Oh, and Pramod Viswanath
  NIPS 2014

• Provable Tensor Factorization with Missing Data
  Prateek Jain and Sewoong Oh
  NIPS 2014, [code]

• Minimax-optimal Inference from Partial Rankings
  Bruce Hajek, Sewoong Oh, and Jiaming Xu
  NIPS 2014

• Learning Mixed Multinomial Logit Model from Ordinal Data
  Sewoong Oh and Devavrat Shah
  NIPS 2014

• Learning Mixtures of Discrete Product Distributions using Spectral Decompositions
  Prateek Jain and Sewoong Oh
  COLT 2014

• The Composition Theorem for Differential Privacy
  Peter Kairouz, Sewoong Oh and Pramod Viswanath
  ICML 2015

• What's your choice? Learning the mixed multi-nomial logit model
  Ammar Ammar, Sewoong Oh, Devavrat Shah, and Luis-Filipe Voloch
  SIGMETRICS (short paper) 2014

• Efficient Crowdsourcing for Multi-class Labeling
  David Karger, Sewoong Oh, and Devavrat Shah
  SIGMETRICS 2013

• Iterative Ranking from Pairwise Comparisons
  Sahand Negahban, Sewoong Oh, and Devavrat Shah
  NIPS 2012, (Spotlight presentation)

• Iterative Learning for Reliable Crowdsourcing Systems
  David R. Karger, Sewoong Oh, and Devavrat Shah
  NIPS 2011, (Oral presentation)

• Budget-optimal Crowdsourcing using Low-rank Matrix Approximations
  David R. Karger, Sewoong Oh, and Devavrat Shah
  Allerton 2011

• Gossip PCA
  Satish Babu Korada, Andrea Montanari, and Sewoong Oh
  ACM SIGMETRICS 2011

• On Positioning via Distributed Matrix Completion
  A. Montanari and S. Oh
  Sensor Array and Multichannel Signal Processing Workshop 2010

• Ultrasound Tomography Calibration using Structured Matrix Completion
  Reza Parhizkar, Amin Karbasi, Sewoong Oh, and Martin Vetterli
  The 20th International Congress on Acoustics, 2010

• Distributed Sensor Network Localization from Local Connectivity: Performance Analysis for the HOP-TERRAIN Algorithm
  Amin Karbasi and Sewoong Oh
  ACM SIGMETRICS 2010, (Kenneth C. Sevcik Outstanding Student Paper Award)

• Sensor Network Localization from Local Connectivity: Performance Analysis for the MDS-MAP Algorithm
  Sewoong Oh, Amin Karbasi, and Andrea Montanari
  Information Theory Workshop 2010

• OptSpace: A Gradient Descent Algorithm on Grassmann Manifold for Matrix Completion
  Raghunandan Keshavan and Sewoong Oh
  Technical report

• Low-rank Matrix Completion with Noisy Observations: a Quantitative Comparison
  Raghunandan Keshavan, Andrea Montanari, and Sewoong Oh
  Allerton 2009

• Matrix Completion from Noisy Entries
  Raghunandan Keshavan, Andrea Montanari, and Sewoong Oh
  NIPS 2009

• Matrix Completion from a Few Entries
  Raghunandan Keshavan, Andrea Montanari, and Sewoong Oh
  ISIT 2009

• Generating Random Tanner-graphs with Large Girth
  Mohsen Bayati, Raghunandan H. Keshavan, Andrea montanari, Sewoong Oh, and Amin Saberi
  Information Theory Workshop 2009, [ code ]

• Learning low rank matrices from O(n) entries
  Raghunandan Keshavan, Andrea Montanari, and Sewoong Oh
  Allerton 2008

• Computing the threshold shift for general channels
  Jeremie Ezri, Andrea Montanari, Sewoong Oh, and Ruediger Urbanke
  ISIT 2008

• The Slope Scaling Parameter for General Channels, Decoders and Ensembles
  Jeremie Ezri, Andrea Montanari, Sewoong Oh, and Ruediger Urbanke
  ISIT 2008

Journal papers

Stochastic optimization on matrices and a graphon McKean–Vlasov limit

Zaid Harchaoui, Sewoong Oh, Soumik Pal, Raghav Somani, Raghavendra Tripathi
Annals of Applied Probability, 2026, Vol. 36, No. 1, 355 - 392.

Sequence-to-sequence translation from mass spectra to peptides with a transformer model

Melih Yilmaz, William E. Fondrie, Wout Bittremieux, Carlo F. Melendez, Rowan Nelson, Varun Ananth, Sewoong Oh, William Stafford Noble
Nature Comunications, 2024, 15(1), pp.6427

Accounting for digestion enzyme bias in Casanovo

Carlo Melendez, Justin Sanders, Melih Yilmaz, Wout Bittremieux, William E Fondrie, Sewoong Oh, William Stafford Noble
Journal of Proteome Research, 2024,

MAUVE Scores for Generative Models: Theory and Practice

Krishna Pillutla, Lang Liu, John Thickstun, Sean Welleck, Swabha Swayamdipta, Rowan Zellers, Sewoong Oh, Yejin Choi, Zaid Harchaoui
Journal of Machine Learning Research, 2023, Volume 24, Issue 1, Article No.: 356, Pages 17105 - 171

Towards a Defense Against Federated Backdoor Attacks Under Continuous Training

Shuaiqi Wang, Jonathan Hayase, Giulia Fanti, Sewoong Oh
Transactions on Machine Learning Research (TMLR), 2023

Machine Learning-Aided Efficient Decoding of Reed-Muller Subcodes

Mohammad Vahid Jamali, Xiyang Liu, Ashok Vardhan Makkuva, Hessam Mahdavifar, Sewoong Oh, Pramod Viswanath
IEEE Transactions on Selected Areas in Information Theory (JSAIT) , 2023

Gradient flows on graphons: existence, convergence, continuity equations

Sewoong Oh, Soumik Pal, Raghav Somani, Raghavendra Tripathi
Journal of Theoretical Probability, 2023,
presented at the NeurIPS 2021 workshop on Optimal Transport and Machine Learning

Evaluating proteomics imputation methods with improved criteria

L Harris, WE Fondrie, S Oh, WS Noble
Journal of Proteome Research, 2023, 22 (11), 3427-3438

Reducing peptide sequence bias in quantitative mass spectrometry data with machine learning

AB Dincer, Y Lu, DK Schweppe, S Oh, WS Noble
Journal of Proteome Research, 2022, 21 (7), 1771-1782

Physical Layer Communication via Deep Learning

Hyeji Kim, Sewoong Oh, Pramod Viswanath
IEEE Transactions on Selected Areas in Information Theory (JSAIT), Vol.1, no.1, pp.5-18, 2020,

LEARN Codes: Inventing Low-latency Codes via Recurrent Neural Networks

Yihan Jiang, Hyeji Kim, Himanshu Asnani, Sreeram Kannan, Sewoong Oh, and Pramod Viswanath
IEEE Transactions on Selected Areas in Information Theory (JSAIT), Vol.1, no.1, pp.207-216, 2020,

PacGAN: The power of two samples in generative adversarial networks

Zinan Lin, Ashish Khetan, Giulia Fanti, Sewoong Oh
IEEE Transactions on Selected Areas in Information Theory (JSAIT), 2 Vol.1, no.1, pp.324-335, 2020, [ code ], [ project page]

Deepcode: Feedback Codes via Deep Learning

Hyeji Kim, Yihan Jiang, Sreeram Kannan, Sewoong Oh, Pramod Viswanath
IEEE Transactions on Selected Areas in Information Theory (JSAIT), Vol.1, no.1, pp.194-206, 2020, [ code by Hyeji Kim ], [ code by Yihan Jiang ]

Spectrum Estimation from a Few Entries

Ashish Khetan, Sewoong Oh
Journal of Machine Learning Research, Vol.20, Issue:21, January 2019

Learning from Comparisons and Choices

Sahand Negahban, Sewoong Oh, Kiran Thekumparampil, and Jiaming Xu,
Journal of Machine Learning Research, Vol.19, Issue:40, pp.1-95, September 2018

Generalized Rank-breaking: Computational and Statistical Tradeoffs

Ashish Khetan, Sewoong Oh
Journal of Machine Learning Research, Vol.19, Issue:28, pp.1-42, September 2018 [bibtex]

Optimality of Belief Propagation for Crowdsourced Classification

Jungseul Ok, Sewoong Oh, Jinwoo Shin, Yung Yi
IEEE Transactions on Information Theory, Vol.64, Issue:9, pp.6127-6138, September 2018,

Demystifying Fixed k-Nearest Neighbor Information Estimators

Weihao Gao, Sewoong Oh, Pramod Viswanath
IEEE Transactions on Information Theory, Vol.64, Issue:8, pp.5629-5661 February 2018, [bibtex]

Breaking the Bandwidth Barrier: Geometrical Adaptive Entropy Estimation

Weihao Gao, Sewoong Oh, Pramod Viswanath
IEEE Transactions on Information Theory, Vol.64, Issue:5, pp.3313-3330, May 2018, [bibtex]

Discovering Potential Correlations via Hypercontractivity

Hyeji Kim, Weihao Gao, Sreeram Kannan, Sewoong Oh, and Pramod Viswanath
Entropy, Vol.19, Issue:11, pp.586, October 2017, [ code ], [bibtex]

Data-driven Rank Breaking for Efficient Rank Aggregation

Ashish Khetan, Sewoong Oh
Journal of Machine Learning Research, Vol.17, no.193, pp.1-54, October 2016 [bibtex]

Hiding the Rumor Source

Giulia Fanti, Peter Kairouz, Sewoong Oh, Kannan Ramchandran, and Pramod Viswanath
IEEE Transactions on Information Theory, Vol.63, Issue:10, pp.6679-6713, October 2017 [bibtex]

Metadata-conscious Anonymous Messaging

Giulia Fanti, Peter Kairouz, Sewoong Oh, Kannan Ramchandran, and Pramod Viswanath
IEEE Transactions on Signal and Information Processing over Networks, Volume: 2, Issue: 4, pp.582 - 594, December 2016

Detecting Sponsored Recommendations

Subhashini Krishnasamy, Rajat Sen, Sewoong Oh, and Sanjay Shakkottai
ACM Transactions on Modeling and Performance Evaluation of Computing Systems, Volume 2, Issue 1, pp.6:1–6:29, November 2016

The Composition Theorem for Differential Privacy

Peter Kairouz, Sewoong Oh and Pramod Viswanath
IEEE Transaction on Information Theory, Volume 63, Issue 6, pp.4037-4049, June 2017 [bibtex]

Extremal Mechanisms for Local Differential Privacy

Peter Kairouz, Sewoong Oh, and Pramod Viswanath
Journal of Machine Learning Research, Volume 17, no.17, pp.1-51, April 2016 [bibtex]

RankCentrality: Ranking from Pair-wise Comparisons

Sahand Negahban, Sewoong Oh, and Devavrat Shah
Operations Research, Vol.65, no.1, pp.266-287, October 2016 [bibtex]

The Staircase Mechanisms in Differential Privacy

Q. Geng, P. Kairouz, S. Oh, and P. Viswanath
Selected Topics in Signal Processing, April 2015

Budget-optimal Task Allocation for Reliable Crowdsourcing Systems

David R. Karger, Sewoong Oh and Devavrat Shah
Operations Research, Volume 62 Issue 1, pp.1-24, January-February 2014 [bibtex]

Robust Localization from Incomplete Local Information

Amin Karbasi and Sewoong Oh
IEEE Transactions on Networking, Vol 21, pp.1131-1144, August 2013, [bibtex]

Calibration using Matrix Completion with Application to Ultrasound Tomography

Reza Parhizkar, Amin Karbai, Sewoong Oh and Martin Vetterli
IEEE Transactions on Signal Processing, Vol 61, pp.4923-4933, October 2013, [bibtex]

Counting with the Crowd

Adam Marcus, David Karger, Samuel Madden, Robert Miller, Sewoong Oh
Journal of the VLDB Endowment, Vol. 6, issue 2, pp.109-120, December 2012, [bibtex]

Matrix Completion from Noisy Entries

Raghunandan Keshavan, Andrea Montanari and Sewoong Oh
Journal of Machine Learning Research, vol. 11, pp.2057-2078, July 2010, [ bibtex , code ]

Matrix Completion from a Few Entries

Raghunandan Keshavan, Andrea Montanari and Sewoong Oh
IEEE Transactions on Information Theory,vol. 56,no. 6, pp.2980-2998, June 2010, [ bibtex , code ]

Dissertation

Matrix Completion: Fundamental Limits and Efficient Algorithms

Ph.D. Dissertation, Stanford Univesiry, December 2010