The Differences Between Blue and NLTK

Is it a standard implementation of blue that people use or are there variations there? Yeah, I would say no. Because of non-standardization, and it just does not happen just in blue. It happens in RUSH score as well, we get variations in our comparisons. And it might be an advantage or disadvantage to the model builder. So it's a totally different dimension that we could talk for an hour, I suppose. But I'm actually really thinking that this is a problem, that we should be really paying attention to, especially when reporting our results.