A Way out of the Odyssey: Analyzing and Combining Recent Insights for LSTMs
This work provides an incremental enhancement for NLP practitioners by offering a better baseline for text classification tasks.
The authors tackled the problem of improving LSTM performance for text classification by analyzing and combining recent insights, resulting in a reliable baseline model with compounding improvements from techniques like Monte Carlo test-time model averaging, average pooling, and residual connections.
LSTMs have become a basic building block for many deep NLP models. In recent years, many improvements and variations have been proposed for deep sequence models in general, and LSTMs in particular. We propose and analyze a series of augmentations and modifications to LSTM networks resulting in improved performance for text classification datasets. We observe compounding improvements on traditional LSTMs using Monte Carlo test-time model averaging, average pooling, and residual connections, along with four other suggested modifications. Our analysis provides a simple, reliable, and high quality baseline model.