Post

@nmw.bsky.social

Strong disagree -this paper is good and I think the extremely talented PhD student that authored it should be proud - she has a bright future ahead of her in science. Pulling ML into a more rigorous paradigm of annotation is what we need much more of - this paper helps that cause.

Laura K. Nelson@lauraknelson.bsky.social

Treading lightly because I'm happy to see data annotation discussions happening anywhere. But I'm also incredibly frustrated that the AI/ML world seems to be happy ignoring literal *decades* of work on this topic. From a recently published paper in AAAI/ACM that has tipped me over into despair fr

15/10/26, 2:48 am0 reposts2 likes

Laura K. Nelson

@lauraknelson.bsky.social

Oh the paper is fantastic! I realize I didn't make that clear. What I'm in despair about is the first sentence in the abstract, which is diagnosing the problem: people in AI have not been engaging with measurement theory. I'm very glad this paper has been published in this venue.

Laura K. Nelson

@lauraknelson.bsky.social

As I clarified later down in the thread bsky.app/profile/laur...

Laura K. Nelson@lauraknelson.bsky.social

The paper itself is quite good. It's basically: hey CS/ML folk, there's this thing called measurement theory in the social sciences, and you should pay attention to it. Yes, excellent! But whyyy was no one paying attention to this massive body of literature before October, 2026?