Orion Weller

promptriever. The first dense retrieval model that can be prompted like an LM

RedditHumorDetection. Code and datasets for the paper "Humor Detection: A Transformer Gets the Last Laugh"

rJokesData. A large scale Humor Dataset, containing more than 550k rated English jokes (LREC'20)

rank1. Test-time compute in information retrieval

FollowIR. FollowIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions

Multilingual-Federated-Learning. Code for the paper "Pretrained Models for Multilingual Federated Learning" at NAACL 2022

humorTranslate. Using Machine Translation to "translate" non-humor into humor. Code for the paper "Humorous Headline Generation via Style Transfer" at FigLang 2020

bert24. Python

MTLvsIFT. Code for the paper "When to Use Multi-Task Learning vs Intermediate Fine-Tuning for Pre-Trained Encoder Transfer Learning"

LM-expansions. When do Generative Query and Document Expansions Fail? A Comprehensive Study Across Methods, Retrievers, and Datasets

according-to. Getting language models to quote from their pre-training data (EACL'24)

NevIR. Negation in Information Retrieval (EACL'24)

configtune. An easy way to tune machine learning hyperparameters (especially for those that use a config file)

mteb-lite. Python

DocumentReadingTime. Code and data from the ACL paper "You Don’t Have Time to Read This: an Exploration of Document-LevelReading Time Prediction"

BYU-Housing-Finder. App for Student Housing - see link to Google Play

GANExperiments. My experiments with GANs

LLaMA-Factory. Unify Efficient Fine-Tuning of 100+ LLMs

tevatron. Tevatron - A flexible toolkit for neural retrieval research and development.

disinformation-defense. Defending Against Misinformation Attacks in Open-Domain Question Answering

LLM-expansions.

ModelComparison. An R Package to return a variety of different model types, complete with hyper-parameter tuning