Conference Papers
2025
- VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data
Thomas Zeng, Shuibai Zhang, Shutong Wu, Christian Classen, Daewon Chae, Ethan Ewer, Minjae Lee, Heeju Kim, Wonjun Kang, Jackson Kunde, Ying Fan, Jungtaek Kim, Hyung Il Koo, Kannan Ramchandran, Dimitris Papailiopoulos, and Kangwook Lee
ICML 2025 (spotlight) | Summary / Github / HuggingFace - Everything Everywhere All at Once: LLMs can In-Context Learn Multiple Tasks in Superposition
Zheyang Xiong, Ziyang Cai, John Cooper, Albert Ge, Vasilis Papageorgiou, Zack Sifakis, Angeliki Giannou, Ziqian Lin, Liu Yang, Saurabh Agarwal, Grigorios Chrysos, Samet Oymak, Kangwook Lee, and Dimitris Papailiopoulos
ICML 2025 (spotlight) - Parameter-Efficient Fine-Tuning of State Space Models
Kevin Galim, Wonjun Kang, Yuchen Zeng, Hyung Il Koo, and Kangwook Lee
ICML 2025 - Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges
Nayoung Lee, Ziyang Cai, Avi Schwarzschild, Kangwook Lee, and Dimitris Papailiopoulos
ICML 2025 - Looped Transformers for Length Generalization
Ying Fan, Yilun Du, Kannan Ramchandran, and Kangwook Lee
ICLR 2025 | Summary | Github - From Artificial Needles to Real Haystacks: Improving Retrieval Capabilities in LLMs by Finetuning on Synthetic Data
Zheyang Xiong, Vasilis Papageorgiou, Kangwook Lee, and Dimitris Papailiopoulos
ICLR 2025 | Summary | Github - Rare-to-Frequent: Unlocking Compositional Generation Power of Diffusion Models on Rare Concepts with LLM Guidance
Dongmin Park, Sebin Kim, Taehong Moon, Minkyu Kim, Kangwook Lee, and Jaewoong Cho
ICLR 2025 (spotlight) | Summary | Github
2024
- Can MLLMs Perform Text-to-Image In-Context Learning?
Yuchen Zeng*, Wonjun Kang*, Yicong Chen, Hyung Il Koo, and Kangwook Lee
COLM 2024 | Summary | Github - Dual Operating Modes of In-Context Learning
Ziqian Lin and Kangwook Lee
ICML 2024 | Summary | Github - Can Mamba Learn How To Learn? A Comparative Study on In-Context Learning Tasks
Jongho Park, Jaeseung Park, Zheyang Xiong, Nayoung Lee, Jaewoong Cho, Samet Oymak, Kangwook Lee, and Dimitris Papailiopoulos
ICML 2024 | Summary | Github - Memorization Capacity for Additive Fine-Tuning with Small ReLU Networks
Jy-yong Sohn, Dohyun Kwon, Seoyeon An, and Kangwook Lee
UAI 2024 | Summary | Github - The Expressive Power of Low-Rank Adaptation
Yuchen Zeng and Kangwook Lee
ICLR 2024 | Summary | Github - Image Clustering Conditioned on Text Criteria
Sehyun Kwon, Jaeseung Park, Minkyu Kim, Jaewoong Cho, Ernest K. Ryu, and Kangwook Lee
ICLR 2024 | Summary | Github - Teaching Arithmetic to Small Transformers
Nayoung Lee, Kartik Sreenivasan, Jason Lee, Kangwook Lee, and Dimitris Papailiopoulos
ICLR 2024 | Summary | Github - Looped Transformers are Better at Learning Learning Algorithms
Liu Yang, Kangwook Lee, Robert D Nowak, and Dimitris Papailiopoulos
ICLR 2024 | Summary | Github
2023
- DPOK: Reinforcement Learning for Fine-tuning Text-to-Image Diffusion Models (code)
Ying Fan, Olivia Watkins, Yuqing Du, Hao Liu, Moonkyung Ryu, Craig Boutilier, Pieter Abbeel, Mohammad Ghavamzadeh, Kangwook Lee, and Kimin Lee
NeurIPS 2023 - Prompted LLMs as Chatbot Modules for Long Open-domain Conversation (code)
Gibbeum Lee, Volker Hartmann, Jongho Park, Dimitris Papailiopoulos, and Kangwook Lee
ACL 2023 (Findings, Short) - Improving Fair Training under Correlation Shifts
Yuji Roh, Kangwook Lee, Steven Euijong Whang, and Changho Suh
ICML 2023 - Optimizing DDPM Sampling with Shortcut Fine-Tuning (code)
Ying Fan and Kangwook Lee
ICML 2023 - Looped Transformers as Programmable Computers (code)
Angeliki Giannou*, Shashank Rajput*, Jy-yong Sohn, Kangwook Lee, Jason D. Lee, and Dimitris Papailiopoulos
ICML 2023 - Federated Learning with Local Fairness Constraints
Yuchen Zeng, Hongxu Chen, and Kangwook Lee
ISIT 2023 - Equal Improvability: A New Fairness Notion Considering the Long-Term Impact (code)
Ozgur Guldogan*, Yuchen Zeng*, Jy-yong Sohn, Ramtin Pedarsani, and Kangwook Lee
ICLR 2023 (Article)
2022
- Online Federated Learning based Object Detection across Autonomous Vehicles in a Virtual World
Shenghong Dai, S M Iftekharul Alam, Ravikumar Balakrishnan, Kangwook Lee, Suman Banerjee, and Nageen Himayat
IEEE CCNC 2023 (Demo) - Utilizing Language-Image Pretraining for Efficient and Robust Bilingual Word Alignment (code)
Tuan Dinh, Jy-yong Sohn, Shashank Rajput, Tim Ossowski, Yifei Ming, Junjie Hu, Dimitris Papailiopoulos, and Kangwook Lee
EMNLP 2022 (Findings) - Score-based generative modeling secretly minimizes the Wasserstein distance (code)
Dohyun Kwon, Ying Fan, and Kangwook Lee
NeurIPS 2022 - LIFT: Language-Interfaced FineTuning for Non-Language Machine Learning Tasks (code)
Tuan Dinh*, Yuchen Zeng*, Ruisu Zhang, Ziqian Lin, Michael Gira, Shashank Rajput, Jy-yong Sohn, Dimitris Papailiopoulos, and Kangwook Lee
NeurIPS 2022 - Rare Gems: Finding Lottery Tickets at Initialization (code)
Kartik Sreenivasan, Jy-yong Sohn, Liu Yang, Matthew Grinde, Aliot Nagle, Hongyi Wang, Kangwook Lee, and Dimitris Papailiopoulos
NeurIPS 2022 - GenLabel: Mixup Relabeling using Generative Models (code)
Jy-yong Sohn, Liang Shang, Hongxu Chen, Jaekyun Moon, Dimitris Papailiopoulos, and Kangwook Lee
ICML 2022 - Breaking Fair Binary Classification with Optimal Flipping Attacks
Changhun Jo, Jy-yong Sohn, and Kangwook Lee
ISIT 2022 (Article) - Permutation-Based SGD: Is Random Optimal?
Shashank Rajput, Kangwook Lee, and Dimitris Papailiopoulos
ICLR 2022
2021
- Sample Selection for Fair and Robust Training (code)
Yuji Roh, Kangwook Lee, Steven Euijong Whang, and Changho Suh
NeurIPS 2021 - Gradient Inversion with Generative Image Prior (code)
Jinwoo Jeon, Jaechang Kim, Kangwook Lee, Sewoong Oh, and Jungseul Ok
NeurIPS 2021 - Coded-InvNet for Resilient Prediction Serving Systems (code)
Tuan Dinh and Kangwook Lee
ICML 2021 (long oral) - Discrete-Valued Latent Preference Matrix Estimation with Graph Side Information
Changhun Jo and Kangwook Lee
ICML 2021 - Accordion: Adaptive Gradient Communication via Critical Learning Regime Identification (code)
Saurabh Agarwal, Hongyi Wang, Kangwook Lee, Shivaram Venkataraman, and Dimitris Papailiopoulos
MLSys 2021 - FairBatch: Batch Selection for Model Fairness (code)
Yuji Roh, Kangwook Lee, Steven Euijong Whang, and Changho Suh
ICLR 2021
2020
- Attack of the Tails: Yes, You Really Can Backdoor Federated Learning (code)
Hongyi Wang, Kartik Sreenivasan, Shashank Rajpu, Harit Vishwakarma, Saurabh Agarwal, Jy-yong Sohn, Kangwook Lee, and Dimitris Papailiopoulos
NeurIPS 2020 - Reprogramming GANs via Input Noise Design (code)
Kangwook Lee, Changho Suh, and Kannan Ramchandran
ECML PKDD 2020 - FR-Train: A mutual information-based approach to fair and robust training (code)
Yuji Roh, Kangwook Lee, Steven Euijong Whang, and Changho Suh
ICML 2020
2019
- Synthesizing Differentially Private Datasets using Random Mixing
Kangwook Lee, Hoon Kim, Kyungmin Lee, Changho Suh, and Kannan Ramchandran
IEEE ISIT 2019 - Crash to Not Crash: Learn to Identify Dangerous Vehicles using a Simulator (site)
Hoon Kim*, Kangwook Lee*, Gyeongjo Hwang, and Changho Suh
AAAI 2019 long oral
2018
- Binary Rating Estimation with Graph Side Information
Kwangjun Ahn, Kangwook Lee, Hyunseung Cha, and Changho Suh
NeurIPS 2018 - On the Joint Recovery of Community Structure and Community Features
Jisang Yoon, Kangwook Lee, and Changho Suh
Allerton Conference on Communication, Control and Computing 2018 - Hierarchical Coding for Distributed Computing
Hyegyeong Park, Kangwook Lee, Jy-yong Sohn, Changho Suh, and Jaekyun Moon
IEEE ISIT 2018 - Straggler-proofing massive-scale distributed matrix multiplication with d-dimensional product codes
Tavor Baharav, Kangwook Lee, Orhan Ocal, and Kannan Ramchandran
IEEE ISIT 2018 - Simulated+Unsupervised Learning With Adaptive Data Generation and Bidirectional Mappings
Kangwook Lee*, Hoon Kim*, and Changho Suh
ICLR 2018 - SGD on Random Mixtures: Private Machine Learning under Data-breach Threats
Kangwook Lee, Kyungmin Lee, Hoon Kim, Changho Suh, and Kannan Ramchandran
SysML 2018 - UberShuffle: Communication-efficient Data Shuffling for SGD via Coding Theory
Jichang Chung, Kangwook Lee, Ramtin Pedarsani, Dimitris Papailiopoulos, and Kannan Ramchandran*
SysML 2018
<= 2017
- Matrix Sparsification for Coded Matrix Multiplication
Geewon Suh, Kangwook Lee, and Changho Suh
Allerton Conference on Communication, Control and Computing 2017 - High-Dimensional Coded Matrix Multiplication
Kangwook Lee, Changho Suh, and Kannan Ramchandran
IEEE ISIT 2017 - Coded Computation for Multicore Setups
Kangwook Lee, Ramtin Pedarsani, Dimitris Papailiopoulos, and Kannan Ramchandran
IEEE ISIT 2017 - Information-theoretic Limits of Subspace Clustering
Kwangjun Ahn, Kangwook Lee, and Changho Suh
IEEE ISIT 2017 - Asynchronous and Noncoherent Neighbor Discovery for the IoT Using Sparse-Graph Codes
Kabir Chandrasekher, Kangwook Lee, Peter Kairouz, Ramtin Pedarsani, and Kannan Ramchandran*
IEEE ICC 2017 - Community Recovery in Hypergraphs
Kwangjun Ahn, Kangwook Lee, and Changho Suh
Allerton Conference on Communication, Control and Computing 2016 - Speeding Up Distributed Machine Learning Using Codes
Kangwook Lee, Maximilian Lam, Ramtin Pedarsani, Dimitris Papailiopoulos, and Kannan Ramchandran*
IEEE ISIT 2016 - SAFFRON: Sparse-Graph Code Framework for Group Testing
Kangwook Lee, Ramtin Pedarsani, and Kannan Ramchandran
IEEE ISIT 2016 - On Scheduling Redundant Requests with Cancellation Overheads
Kangwook Lee, Ramtin Pedarsani, and Kannan Ramchandran
Allerton Conference on Communication, Control and Computing 2015 - Sparse Covariance Estimation Based on Sparse-Graph Codes
Ramtin Pedarsani, Kangwook Lee, and Kannan Ramchandran
Allerton Conference on Communication, Control and Computing 2015 - Fast and Robust Compressive Phase Retrieval with Sparse-Graph Codes
Dong Yin, Kangwook Lee, Ramtin Pedarsani, and Kannan Ramchandran
IEEE ISIT 2015 - Capacity-Approaching PhaseCode for Low-Complexity Compressive Phase Retrieval
Ramtin Pedarsani, Kangwook Lee, and Kannan Ramchandran
IEEE ISIT 2015 - PhaseCode: Fast and Efficient Compressive Phase Retrieval based on Sparse-Graph-Codes
Ramtin Pedarsani, Kangwook Lee, and Kannan Ramchandran
Allerton Conference on Communication, Control and Computing 2014 - The MDS Queue: Analysing the Latency Performance of Codes
Nihar Shah, Kangwook Lee, and Kannan Ramchandran
IEEE ISIT 2014 - When Do Redundant Requests Reduce Latency?
Nihar Shah, Kangwook Lee, and Kannan Ramchandran
Allerton Conference on Communication, Control and Computing 2013 - A VoD System for Massively Scaled, Heterogeneous Environments: Design and Implementation (code)
Kangwook Lee, Lisa Yan, Abhay Parekh, and Kannan Ramchandran
IEEE MASCOTS 2013 Best Paper Award finalist - An Optimized Distributed Video-on-Demand Streaming System: Theory and Design (code)
Kangwook Lee, Hao Zhang, Ziyu Shao, Minghua Chen, Abhay Parekh, and Kannan Ramchandran
Allerton Conference on Communication, Control and Computing 2012 - Codes for a Distributed Caching based Video-On-Demand System
Sameer Pawar, Salim Rouayheb, Hao Zhang, Kangwook Lee, and Kannan Ramchandran
Asilomar Conference on Signals, Systems, and Computers 2011 - Experiment evaluation of optimal CSMA
Bruno Nardelli, Jinsung Lee, Kangwook Lee, Yung Yi, Song Chong, Edward Knightly, and Mung Chiang
IEEE INFOCOM 2011
Workshop Papers (last updated in 2023)
- Image Clustering Conditioned on Text Criteria
Sehyun Kwon, Jaeseung Park, Minkyu Kim, Jaewoong Cho, Ernest K. Ryu, and Kangwook Lee
NeurIPS 2023 Workshop on Robustness of Few-shot and Zero-shot Learning in Foundation Models (R0-FOMO) - Coded Prompts for Large Language Models
Ziqian Lin, Yicong Chen, Yuchen Zeng, and Kangwook Lee
NeurIPS 2023 Workshop on Robustness of Few-shot and Zero-shot Learning in Foundation Models (R0-FOMO) - Zero-shot Improvement of Object Counting with CLIP
Ruisu Zhang, Yicong Chen, and Kangwook Lee
NeurIPS 2023 Workshop on Robustness of Few-shot and Zero-shot Learning in Foundation Models (R0-FOMO) - The Expressive Power of Low-Rank Adaptation [Github]
Yuchen Zeng and Kangwook Lee
NeurIPS 2023 Workshop on Optimization for Machine Learning (OPT 2023) - Teaching Arithmetic to Small Transformers
Nayoung Lee, Kartik Sreenivasan, Jason Lee, Kangwook Lee, and Dimitris Papailiopoulos
NeurIPS 2023 Workshop on Mathematical Reasoning and AI - Outlier-Robust Group Inference via Gradient Space Clustering
Yuchen Zeng, Kristjan Greenewald, Luann Jung, Kangwook Lee, Justin Solomon, Mikhail Yurochkin
NeurIPS 2023 Workshop on Distribution Shifts (DistShift) - Super-Resolution Emulation of Large Cosmological Fields with a 3D Conditional Diffusion Model
Adam Rouhiainen, Michael Gira, Gary Shiu, Kangwook Lee, and Moritz Münchmeyer
NeurIPS 2023 Workshop on Machine Learning and the Physical Sciences - Predictive Pipelined Decoding: A Compute-Latency Trade-off for Exact LLM Decoding
Seongjun Yang, Gibbeum Lee, Jaewoong Cho, Dimitris Papailiopoulos, and Kangwook Lee
ICML 2023 Workshop on Efficient Systems for Foundation Models - Looped Transformers are Better at Learning Learning Algorithms
Liu Yang, Kangwook Lee, Robert D Nowak, and Dimitris Papailiopoulos
ICML 2023 Workshop on Efficient Systems for Foundation Models - A Representer Theorem for Vector-Valued Neural Networks: Insights on Weight Decay Training and Widths of Deep Neural Networks
Joseph Shenouda, Rahul Parhi, Kangwook Lee, and Robert D Nowak
ICML 2023 Workshop on Duality Principles for Modern Machine Learning - Teaching Arithmetic to Small Transformers
Nayoung Lee, Kartik Sreenivasan, Jason Lee, Kangwook Lee, and Dimitris Papailiopoulos
ICML 2023 Workshop on Neural Conversational AI Workshop - FedGP: Buffer-based Gradient Projection for Continual Federated Learning
Shenghong Dai, Bryce Yicong Chen, Jy-yong Sohn, S M Iftekharul Alam, Ravikumar Balakrishnan, Suman Banerjee, Nageen Himayat, Kangwook Lee
MLSys-FLSys 2023 Best Paper Award - Looped Transformers as Programmable Computers
Angeliki Giannou, Shashank Rajput, Jy-yong Sohn, Kangwook Lee, Jason D. Lee, and Dimitris Papailiopoulos
ICLR 2023 Workshop on Mathematical and Empirical Understanding of Foundation Models - Mini-Batch Optimization of Contrastive Loss
Kartik Sreenivasan, Keon Lee, Jeong-Gwan Lee, Anna Lee, Jaewoong Cho, Jy-yong Sohn, Dimitris Papailiopoulos, and Kangwook Lee
ICLR 2023 Workshop on Mathematical and Empirical Understanding of Foundation Models - Active Learning is a Strong Baseline for Data Subset Selection
Dongmin Park, Dimitris Papailiopoulos, and Kangwook Lee
NeurIPS 2022 HITY Workshop - A Better Way to Decay: Proximal Gradient Training Algorithms for Neural Nets
Liu Yang, Jifan Zhang, Joseph Shenouda, Dimitris Papailiopoulos, Kangwook Lee, and Robert D. Nowak
NeurIPS 2022 OPT Workshop - Super Seeds: extreme model compression by trading off storage with computation
Nayoung Lee*, Shashank Rajpu*, Jy-yong Sohn, Hongyi Wang, Aliot Nagle, Eric P. Xing, Kangwook Lee, and Dimitris Papailiopoulos
ICML 2022 Workshop on Updatable Machine Learning (UpML 2022) - Improved Input Reprogramming for GAN Conditioning
Tuan Dinh, Daewon Seo, Zhixu Du, Liang Shang, and Kangwook Lee
ICML 2022 Workshop on Updatable Machine Learning (UpML 2022) - Improving Fairness via Federated Learning
Yuchen Zeng, Hongxu Chen, and Kangwook Lee
MLSys-CrossFL 2022 - Dynamic Decentralized Federated Learning
Shenghong Dai, Kangwook Lee, and Suman Banerjee
MLSys-CrossFL 2022 - Debiasing Pre-Trained Language Models via Efficient Fine-tuning
Michael Gira, Ruisu Zhang, and Kangwook Lee
ACL 2022 Workshop on Language Technology for Equality, Diversity, Inclusion - Federated Unsupervised Clustering with Generative Models
Jichang Chung, Kangwook Lee, and Kannan Ramchandran
AAAI 2022 Workshop on Federated Learning - Improving Fairness via Federated Learning
Yuchen Zeng, Hongxu Chen, and Kangwook Lee
AAAI 2022 Workshop on Federated Learning - Gradient Inversion with Generative Image Prior
Jinwoo Jeon, Jaechang Kim, Kangwook Lee, Sewoong Oh, and Jungseul Ok
ICML 2021 Workshop on Federated Learning for User Privacy and Data Confidentiality - Empirical Study on the Effective VC Dimension of Low-rank Neural Networks
Daewon Seo, Hongyi Wang, Dimitris Papailiopoulos, and Kangwook Lee
ICML 2020 Workshop on Overparameterization: Pitfalls & Opportunities - GAN-mixup: Augmenting Across Data Manifolds for Improved Robustness
Jy-yong Sohn, Kangwook Lee, Jaekyun Moon, and Dimitris Papailiopoulos
ICML 2020 Workshop on Uncertainty & Robustness in Deep Learning - Improving Model Robustness via Automatically Incorporating Self-supervision Tasks
Dongwha Kim, Kangwook Lee, and Changho Suh
NeurIPS 2019 Workshop on Meta-Learning (MetaLearn 2019) - SGD on Random Mixtures: Private Machine Learning under Data-breach Threats
Kangwook Lee, Kyungmin Lee, Hoon Kim, Changho Suh, and Kannan Ramchandran
ICLR 2018 Workshop - UberShuffle: Communication-efficient Data Shuffling for SGD via Coding Theory
Jichang Chung, Kangwook Lee, Ramtin Pedarsani, Dimitris Papailiopoulos, and Kannan Ramchandran*
NIPS 2017 Workshop on Machine Learning Systems - Crash to not crash: Playing video games to predict vehicle collisions
Kangwook Lee, Hoon Kim, and Changho Suh
ICML 2017 Workshop on Machine Learning for Autonomous Vehicles - Large-scale and Interpretable Collaborative Filtering for Educational Data
Kangwook Lee, Jichang Chung, and Changho Suh
KDD 2017 Workshop on Advancing Education with Data - Machine Learning Approaches for Learning Analytics: Collaborative Filtering or Regression With Experts?
Kangwook Lee, Jichang Chung, Youngmin Cha, and Changho Suh
NIPS 2016 Workshop on Machine Learning for Education - Speeding Up Distributed Machine Learning Using Codes
Kangwook Lee, Maximilian Lam, Ramtin Pedarsani, Dimitris Papailiopoulos, and Kannan Ramchandran*
NIPS 2015 Workshop on Machine Learning Systems