Can preprocessing in time-series data (e.g. deseasonaliation or detrending) helps create better forecasting model?

I am reading a paper that mentions the following.

Recent developments have been mainly around preprocessing techniques such as deseasonalization and detrending to supplement the NN's learning process,....

How do the above mentioned preprocessing methods help to learn the ML-model when it seems counter-intuitive to me?

Topic forecasting lstm neural-network time-series

Category Data Science

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.