The academics is rife with intentionally misleading results and findings that aren't actually innovative or informative. Worse yet, providing source code and a

The academics is rife with intentionally misleading results and findings that aren't actually innovative or informative. Worse yet, providing source code and a way to repeat the results is rare, and what good is a research paper with results that aren't repeatable? ML has a bright future, but it's very hard for me to take the academic research seriously. That's why it seems like few care about anything beyond making new models and bolstering their resume by fudging the results.