Kawin

dataset_difficulty. "Understanding Dataset Difficulty with V-Usable Information" (ICML 2022, outstanding paper)

usif. Implementation of unsupervised smoothed inverse frequency (Best Paper, Repl4NLP @ ACL 2018)

contextual. How Contextual are Contextualized Word Representations?

kawine.github.io. HTML

color_analysis. An analysis of color in 19th century literature. Joint project with Stanford Literary Lab.

steamshp. Code for training SteamSHP models.

atgan. GAN for adversarial attacks; CSC2541 project.

HALOs. A library with extensible implementations of DPO, KTO, PPO, ORPO, and other human-aware loss functions (HALOs).