Picture for Kangwook Lee

Kangwook Lee

Vector-Valued Variation Spaces and Width Bounds for DNNs: Insights on Weight Decay Regularization

May 25, 2023
Joseph Shenouda, Rahul Parhi, Kangwook Lee, Robert D. Nowak

Access Paper or Ask Questions

Prompted LLMs as Chatbot Modules for Long Open-domain Conversation

May 08, 2023
Gibbeum Lee, Volker Hartmann, Jongho Park, Dimitris Papailiopoulos, Kangwook Lee

Figure 1 for Prompted LLMs as Chatbot Modules for Long Open-domain Conversation

Figure 2 for Prompted LLMs as Chatbot Modules for Long Open-domain Conversation

Figure 3 for Prompted LLMs as Chatbot Modules for Long Open-domain Conversation

Figure 4 for Prompted LLMs as Chatbot Modules for Long Open-domain Conversation

* Accepted to the Findings of ACL2023. The camera-ready version with additional experimental results will be uploaded

Access Paper or Ask Questions

Optimizing DDPM Sampling with Shortcut Fine-Tuning

Feb 01, 2023
Ying Fan, Kangwook Lee

Figure 1 for Optimizing DDPM Sampling with Shortcut Fine-Tuning

Figure 2 for Optimizing DDPM Sampling with Shortcut Fine-Tuning

Figure 3 for Optimizing DDPM Sampling with Shortcut Fine-Tuning

Figure 4 for Optimizing DDPM Sampling with Shortcut Fine-Tuning

Access Paper or Ask Questions

Score-based Generative Modeling Secretly Minimizes the Wasserstein Distance

Dec 13, 2022
Dohyun Kwon, Ying Fan, Kangwook Lee

Figure 1 for Score-based Generative Modeling Secretly Minimizes the Wasserstein Distance

Figure 2 for Score-based Generative Modeling Secretly Minimizes the Wasserstein Distance

Figure 3 for Score-based Generative Modeling Secretly Minimizes the Wasserstein Distance

Figure 4 for Score-based Generative Modeling Secretly Minimizes the Wasserstein Distance

* 36th Conference on Neural Information Processing Systems (NeurIPS 2022)

Access Paper or Ask Questions

Outlier-Robust Group Inference via Gradient Space Clustering

Oct 13, 2022
Yuchen Zeng, Kristjan Greenewald, Kangwook Lee, Justin Solomon, Mikhail Yurochkin

Figure 1 for Outlier-Robust Group Inference via Gradient Space Clustering

Figure 2 for Outlier-Robust Group Inference via Gradient Space Clustering

Figure 3 for Outlier-Robust Group Inference via Gradient Space Clustering

Figure 4 for Outlier-Robust Group Inference via Gradient Space Clustering

* 17 pages, 6 tables, 8 figures

Access Paper or Ask Questions

Equal Improvability: A New Fairness Notion Considering the Long-term Impact

Oct 13, 2022
Ozgur Guldogan, Yuchen Zeng, Jy-yong Sohn, Ramtin Pedarsani, Kangwook Lee

Figure 1 for Equal Improvability: A New Fairness Notion Considering the Long-term Impact

Figure 2 for Equal Improvability: A New Fairness Notion Considering the Long-term Impact

Figure 3 for Equal Improvability: A New Fairness Notion Considering the Long-term Impact

Figure 4 for Equal Improvability: A New Fairness Notion Considering the Long-term Impact

* Codes are available in a GitHub repository, see https://github.com/guldoganozgur/ei_fairness. 19 pages, 5 figures, 4 tables

Access Paper or Ask Questions