zdxdsw's repositories
inductive_counting_with_LMs
This work provides extensive empirical results on training LMs to count. We find that while traditional RNNs trivially achieve inductive counting, Transformers have to rely on positional embeddings to count out-of-domain. Modern RNNs (e.g. rwkv, mamba) also largely underperform traditional RNNs in generalizing counting inductively.
Automated-Story-Generation-with-BERT
Summer Research in Entertainment Intelligence Lab, Gatech, 2019
skewed_relations_T2I
Codebase for the paper "Skews in the Phenomenon Space Hinder Generalization in Text-to-Image Generation"
COMP3111-Spring2018-ProjectBaseCode
A base code for COMP3111 course project (Spring 2018)
GT-CS3600
2019 Spring - Artificial Intelligence
GT-CS4460
2019 Spring - Information Visualization
GT-CS4641
2019 Spring - Machine Learning
HKUST-COMP2012-Programming-Assignment
2017 Fall - Intro to OOP and Data Structures
HKUST-COMP4651-Course-Project
2018 Fall - Cloud Computing
HKUST-COMP4901K-Project3
Language Model Course Project - 2018 Fall - Machine Learning for NLP
Pytorch-UNet
Pytorch implementation of the U-Net for image semantic segmentation, with dense CRF post-processing
refer_lichengunc
python3 version of https://github.com/lichengunc/refer (their old code was written in python2.7)
Wikipedia
A Pythonic wrapper for the Wikipedia API
zdxdsw.github.io
Personal Site