Can LLMs figure out who you are from your anonymous posts?
From a handful of comments, LLMs can infer where you live, what you do, and your interests; then search for you on the web.
New 📄 w/ @SimonLermenAI, @joshua_swans, @AerniMichael, Nicholas Carlini, @florian_tramer 🧵
our paper on data mixing for LMs is out!
while building Olmo 3, we saw gaps between data mixing literature and real practice
🐠choosing proxy size, # runs, sampling, regression, constraints..
🐟data shifts during LM dev: can we reuse past experiments?
Olmix tackles them all!
aclanthology.org/2025.finding...
If you're attending #EMNLP2025, we'll be presenting virtually in Gather Session 1 on Nov 5 at 4pm PT. Come say hello!
w/ the wonderful:
@mellymeldubs.bsky.social
Anna Preus,
@mariaa.bsky.social
Paper: arxiv.org/abs/2510.16713
Code/Data: github.com/darthbhyrava/wisp
Dash: poetry.darthbhyrava.com
What if a single model could recognize an author's writing style no matter what language they wrote in? 🌍✍️ Our new #EMNLP2025 paper explores multilingual authorship representation, showing how training across 36 languages can sharpen stylistic signals and reduce topic bias.
👇🧵
New opinion paper out with Esther Ploeger (Aalborg University): We Need to Measure Data Diversity in NLP — Better and Broader at #EMNLP2025 (main) aclanthology.org/2025.emnlp-m...
aclanthology.org/2025.emnlp-m...
aclanthology.org/2025.emnlp-m...
aclanthology.org/2025.emnlp-m...
aclanthology.org/2025.finding...
aclanthology.org/2025.emnlp-m...
aclanthology.org/2025.finding...
aclanthology.org/2025.emnlp-m...
aclanthology.org/2025.emnlp-m...
aclanthology.org/2025.emnlp-m...
Lot's of exciting work on linguistic style this year at #EMNLP2025 #EMNLP! Including work on machine-text detection, authorship representation and more
🧵 with anthology links below
📣 with an open call to everyone to add style work that's missing
I have a new blog post about the so-called “tokenizer-free” approach to language modeling and why it’s not tokenizer-free at all. I also talk about why people hate tokenizers so much!
I successfully defended my PhD in Dutch fashion and required a PhD certificate in Latin. Thank you to the amazing people that got me here, a.o. @dongng.bsky.social and the ones I blur here.
Come join next Wednesday if you want to rant about society's love-hate relationship with LLMs!
one of the other entrances was closed off yesterday, increasing my commute from front door to office by another 10 minutes
Is this the Dutch budget cuts or does utrecht uni really not want me to come to the office? My highlight is the door that has been broken for weeks, with the only change being a laminated piece of paper saying I should enter uni maze through two other buildings.
Tussen Mönchengladbach en Venlo rijden geen treinen. De dienstregeling wordt gehandhaafd door een bus. De bus codeswitcht: een monitor waarop staat „de bus hält“.
Tussen Mönchengladbach en Venlo rijden geen treinen. De dienstregeling wordt gehandhaafd door een bus. De bus codeswitcht
work with and by @yupeidu.bsky.social
Tokenization is Sensitive to Language Variation arxiv.org/abs/2502.15343
Disentangling the Roles of Representation and Selection in Data Pruning. arxiv.org/abs/2507.03648
On Support Samples of Next Word Prediction. arxiv.org/abs/2506.04047
VAQUUM: Are Vague Quantifiers Grounded in Visual Data? arxiv.org/pdf/2502.11874
Utrecht is back from #ACL2025! We had a blast.
I should have posted this before but here are some papers from people in our group that were presented at ACL.
I'm sadly not at #ACL2025, but the work on tokenization seem to continue to explode. Here are the tokenization related papers I could find, in no particular order. Let me know if I missed any.
Since people at #ACL2025 are very interested in tokenization, a reminder to join the discussion on discord set up by @mcognetta.bsky.social
Anyone tried the kiss the cook lunch place at #ACL2025?
I think accepted
I will present our #ACL2025 paper Tokenization is Sensitive to Language Variation in the poster session after Tuesday's keynote, 10.30 - 12.00 in Hall 4/5
@philipwitti.bsky.social will be presenting our paper "Tokenisation is NP-Complete" at #ACL2025 😁 Come to the language modelling 2 session (Wednesday morning, 9h~10h30) to learn more about how challenging tokenisation can be!
We are presenting this paper at #ACL2025 😁 Find us at poster session 4 (Wednesday morning, 11h~12h30) to learn more about tokenisation bias!
Im at #ACL2025 this week.
Happy to chat about measuring linguistic style, data diversity, creating synthetic data for analyzing (L)LMs, authorship attribution, paraphrases and tokenizers.
Let’s chat if you’re around
The #ACL2025 #ACL2025NLP feed is up and running! It matches both hashtags and any posts from or mentions of @aclmeeting.bsky.social
Pin it to your home 📌 and enjoy!
bsky.app/profile/did:...
Who's presenting on subjectivity in annotation (human label variation, learning from disagreement, perspectivism) at #ACL2025?
papers by e.g. @liweijiang.bsky.social @tiancheng.bsky.social @gabriellalapesa.bsky.social @romanklinger.de
keynote @verenarieser.bsky.social
link to full list below ⤵️
I love it.
