Skip to content

Latest commit

 

History

History
10 lines (7 loc) · 492 Bytes

README.md

File metadata and controls

10 lines (7 loc) · 492 Bytes

The Surprising Performance of Simple Baselines for Misinformation Detection

Kellin Pelrine*, Jacob Danovitch*, Reihaneh Rabbany

Paper: https://arxiv.org/pdf/2104.06952.pdf

For parallelization, these are designed to be run as wandb sweeps on a Slurm cluster using the files with _job affix. Please set environment variables before running: WANDB_ENTITY WANDB_PROJECT

To run a single model, extract the allennlp command from the corresponding file (e.g., main_models.py) and run manually.