Scott Geng

realtalk. The official implementation of the paper "Affective Faces for Goal-Driven Dyadic Communication."

delta_learning. Code release for the paper "The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains"

unmet-promise. Code repository for the paper "The Unmet Promise of Synthetic Training Images: Using Retrieved Real Images Performs Better"

lectscript. Natural language processing for Zoom lectures

nand2tetris. my project files for the nand2tetris course (https://www.nand2tetris.org/)