(2/3) Toxicity detection is shaped by norms, identity, & context, which existing approaches overlook. Enter MODELCITIZENS: a new dataset designed to address this.
✔️ 6.8K posts, 40K annotations across diverse groups
✔️ Context-augmented scenarios
✔️ New fine-tuned models that beat GPT-4o-mini by 5.5%
(1/3) Please check out our new paper with @skgabrie.bsky.social and her amazing students, to appear in #EMNLP2025!
(🚨 Offensive Content Warning)
arxiv.org/abs/2507.05455
I'm not personally attached to the generative linguistics apparatus per se, but I was asked by the journal to write this paper as a response to another paper, and that paper is primarily opining about the possible "end of (generative) linguistics as we know it."
I didn't say that social relevance will guarantee generative linguistics's survival (note that there is a subtle difference between "theoretical" and "generative"), but rather that social irrelevance will likely guarantee its demise.
I'm glad you liked it! (I am the author)
There are a couple of points of incommensurability between your reaction and my intentions in writing this piece, which I'll explain below.
I keep thinking "Bluesky" is a Slavic patronymic
💬 Have you or a loved one compared LM probabilities to human linguistic acceptability judgments? You may be overcompensating for the effect of frequency and length!
🌟 In our new paper, we rethink how we should be controlling for these factors 🧵:
