News for hackers. Stuff that matters.

A comment on Bad numbers in the “gzip beats BERT” paper?. Parent comment

Re:Bad numbers in the “gzip beats BERT” paper? by thomasahle on Tuesday July 18, 2023 @02:55AM (#36768793)
I don't think it's "customary" to use OSF in machine learning. At least I had never heard of it. It looks great though, I'm glad other fields are figuring this out. Hopefully it will be taken up by the ML field too.
[ Parent | on HN ]
Re:Bad numbers in the “gzip beats BERT” paper? by scrum-treats on Tuesday July 18, 2023 @11:49AM (#36774414)
Really? Perhaps it was just my department (interdisciplinary). We were strongly encouraged to put our work up for review via OSF, including ML. This promotes transparency, collaboration, and verifiable replication.

Why? You can't trust "peer review" from a journal or a conference alone for the verified stamp of "correctness." Why? Bias and ego; humans are human. So, open it up to the hive, encourage people to swarm.

[ Parent | on HN ]