no code implementations • EMNLP 2021 • Alexander Spangher, Jonathan May, Sz-Rung Shiang, Lingjia Deng
As labeling schemas evolve over time, small differences can render datasets following older schemas unusable.
Ranked #1 on Text Classification on NewsDiscourse
no code implementations • 2 Jan 2021 • Alexander Spangher, Jonathan May, Sz-Rung Shiang, Lingjia Deng
Small class-imbalanced datasets, common in many high-level semantic tasks like discourse analysis, present a particular challenge to current deep-learning architectures.