Anna Wegmann
@annawegmann.bsky.social
Postdoctoral Researcher at Utrecht University | Including different styles in NLP | she/her
https://annawegmann.github.io/
https://annawegmann.github.io/
Pinned
Anna Wegmann
@annawegmann.bsky.social
· Nov 19
Measure the style of your texts using our popular style embedding model huggingface.co/AnnaWegmann/...
Reposted by Anna Wegmann
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
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
October 31, 2025 at 3:36 PM
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
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
Reposted by Anna Wegmann
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.
👇🧵
👇🧵
November 6, 2025 at 5:42 AM
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.
👇🧵
👇🧵
Reposted by Anna Wegmann
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...
We Need to Measure Data Diversity in NLP — Better and Broader
Dong Nguyen, Esther Ploeger. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
aclanthology.org
November 4, 2025 at 3:43 PM
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...
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
🧵 with anthology links below
📣 with an open call to everyone to add style work that's missing
November 4, 2025 at 1:42 PM
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
🧵 with anthology links below
📣 with an open call to everyone to add style work that's missing
Reposted by Anna Wegmann
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!
September 25, 2025 at 3:14 PM
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.
October 22, 2025 at 2:20 PM
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.
Reposted by Anna Wegmann
Come join next Wednesday if you want to rant about society's love-hate relationship with LLMs!
🚨 Upcoming: "Large Language Models for Social Research: Potentials and Challenges"
👤 Indira Sen (University of Mannheim)
🗓️ Wed, October 22, 13:45-15:15 CET
📺 Register for the live stream: us02web.zoom.us/meeting/regi...
🔗 socialsciencedatalab.mzes.uni-mannheim.de/page/events/
👤 Indira Sen (University of Mannheim)
🗓️ Wed, October 22, 13:45-15:15 CET
📺 Register for the live stream: us02web.zoom.us/meeting/regi...
🔗 socialsciencedatalab.mzes.uni-mannheim.de/page/events/
![Large Language Models for Social Research: Potentials and Challenges
Hybrid event [A5, 6, Room A231 + Zoom]
Oktober 22, 2025, 13:45-15:15
Abstract
Large Language Models (LLMs) have the potential to revolutionize the social sciences—for example, by accelerating content analysis or enabling realistic social simulations. In this workshop, I will discuss how LLMs can be applied and audited for social science applications, including the generation of synthetic survey responses and content analysis. I will also address how biases in LLMs can hinder these applications and explore ways to better surface and understand these biases. Finally, I will present a hands-on use case demonstrating how LLMs can be guided using demographic personas for both content analysis and simulated surveys.
Presenter(s)
Indira Sen is a Junior Faculty member at the University of Mannheim’s Business School in the Chair of Data Science for the Social and Economic Sciences. Her work lies at the intersection of NLP and Computational Social Science, specifically in developing and evaluating representative and equitable language technology, including Large Language Models.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:oizrkjzyr6vekl76tjctghcl/bafkreigj6rty2hy4hgawqzq34a24si45dxzutezuphif3hprnl5amxldoy@jpeg)
October 16, 2025 at 9:32 AM
Come join next Wednesday if you want to rant about society's love-hate relationship with LLMs!
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.
August 20, 2025 at 6:28 AM
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.
Reposted by Anna Wegmann
Tussen Mönchengladbach en Venlo rijden geen treinen. De dienstregeling wordt gehandhaafd door een bus. De bus codeswitcht
August 1, 2025 at 8:48 AM
Tussen Mönchengladbach en Venlo rijden geen treinen. De dienstregeling wordt gehandhaafd door een bus. De bus codeswitcht
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 should have posted this before but here are some papers from people in our group that were presented at ACL.
August 5, 2025 at 3:37 PM
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 should have posted this before but here are some papers from people in our group that were presented at ACL.
Reposted by Anna Wegmann
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.
July 30, 2025 at 2:03 PM
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
July 29, 2025 at 12:52 PM
Since people at #ACL2025 are very interested in tokenization, a reminder to join the discussion on discord set up by @mcognetta.bsky.social
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
Wanna do some authorship attribution? Chances are what tokenizer you use matters.
Tokenization is Sensitive to Language Variation, probably, more investigation necessary...
📄 ACL Findings paper: arxiv.org/pdf/2502.15343
🧑🏫 @dongng.bsky.social @davidjurgens.bsky.social and myself
See you at ACL!
Tokenization is Sensitive to Language Variation, probably, more investigation necessary...
📄 ACL Findings paper: arxiv.org/pdf/2502.15343
🧑🏫 @dongng.bsky.social @davidjurgens.bsky.social and myself
See you at ACL!
July 28, 2025 at 5:43 AM
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
Reposted by Anna Wegmann
@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!
BPE is a greedy method to find a tokeniser which maximises compression! Why don't we try to find properly optimal tokenisers instead? Well, it seems this is a pretty difficult—in fact, NP-complete—problem!🤯
New paper + @philipwitti.bsky.social
@gregorbachmann.bsky.social :) arxiv.org/abs/2412.15210
New paper + @philipwitti.bsky.social
@gregorbachmann.bsky.social :) arxiv.org/abs/2412.15210
Tokenisation is NP-Complete
In this work, we prove the NP-completeness of two variants of tokenisation, defined as the problem of compressing a dataset to at most $δ$ symbols by either finding a vocabulary directly (direct token...
arxiv.org
July 27, 2025 at 9:41 AM
@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!
Reposted by Anna Wegmann
We are presenting this paper at #ACL2025 😁 Find us at poster session 4 (Wednesday morning, 11h~12h30) to learn more about tokenisation bias!
A string may get 17 times less probability if tokenised as two symbols (e.g., ⟨he, llo⟩) than as one (e.g., ⟨hello⟩)—by an LM trained from scratch in each situation! Our new ACL paper proposes an observational method to estimate this causal effect! Longer thread soon!
July 27, 2025 at 11:59 AM
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
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
July 27, 2025 at 5:30 PM
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
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
Reposted by Anna Wegmann
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:...
Pin it to your home 📌 and enjoy!
bsky.app/profile/did:...
July 17, 2025 at 11:15 AM
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:...
Pin it to your home 📌 and enjoy!
bsky.app/profile/did:...
Reposted by Anna Wegmann
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 ⤵️
papers by e.g. @liweijiang.bsky.social @tiancheng.bsky.social @gabriellalapesa.bsky.social @romanklinger.de
keynote @verenarieser.bsky.social
link to full list below ⤵️
July 24, 2025 at 4:58 PM
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 ⤵️
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.
We are excited that Scholar Inbox is supporting ACL 2025
@aclmeeting.bsky.social with personalized conference programs this year for the first time!
www.scholar-inbox.com/conference/a...
It would be great if you could widely share this news within the NLP community. See you next week in Vienna!
@aclmeeting.bsky.social with personalized conference programs this year for the first time!
www.scholar-inbox.com/conference/a...
It would be great if you could widely share this news within the NLP community. See you next week in Vienna!
July 24, 2025 at 3:48 PM
I love it.
Reposted by Anna Wegmann
I'll be attending ACL 2025 in Vienna! Looking forward to seeing people there!😊🇦🇹 We are going to present 'LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks' aclanthology.org/2025.acl-sho... #acl2025 #acl2025nlp
July 24, 2025 at 12:34 PM
I'll be attending ACL 2025 in Vienna! Looking forward to seeing people there!😊🇦🇹 We are going to present 'LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks' aclanthology.org/2025.acl-sho... #acl2025 #acl2025nlp
How come the @aclmeeting.bsky.social underline page was set to release July 20 last Friday and now promises access only on the 24th?
Access to papers and videos remains evasive less than a week before the conference.
Access to papers and videos remains evasive less than a week before the conference.
July 22, 2025 at 12:34 PM
How come the @aclmeeting.bsky.social underline page was set to release July 20 last Friday and now promises access only on the 24th?
Access to papers and videos remains evasive less than a week before the conference.
Access to papers and videos remains evasive less than a week before the conference.
Reposted by Anna Wegmann
Do LLMs represent the people they're supposed simulate or provide personalized assistance to?
We review the current literature in our #ACL2025 Findings paper and investigating what researchers conclude about the demographic representativeness of LLMs:
osf.io/preprints/so...
1/
We review the current literature in our #ACL2025 Findings paper and investigating what researchers conclude about the demographic representativeness of LLMs:
osf.io/preprints/so...
1/
July 21, 2025 at 10:12 AM
Do LLMs represent the people they're supposed simulate or provide personalized assistance to?
We review the current literature in our #ACL2025 Findings paper and investigating what researchers conclude about the demographic representativeness of LLMs:
osf.io/preprints/so...
1/
We review the current literature in our #ACL2025 Findings paper and investigating what researchers conclude about the demographic representativeness of LLMs:
osf.io/preprints/so...
1/
Reposted by Anna Wegmann
#ic2s2 I’ll be talking about this paper in one hour in Vingen 1+2!
New preprint!
Metaphors shape how people understand politics, but measuring them (& their real-world effects) is hard.
We develop a new method to measure metaphor & use it to study dehumanizing metaphor in 400K immigration tweets Link: bit.ly/4i3PGm3
#NLP #NLProc #polisky #polcom #compsocialsci
🐦🐦
Metaphors shape how people understand politics, but measuring them (& their real-world effects) is hard.
We develop a new method to measure metaphor & use it to study dehumanizing metaphor in 400K immigration tweets Link: bit.ly/4i3PGm3
#NLP #NLProc #polisky #polcom #compsocialsci
🐦🐦
July 22, 2025 at 8:09 AM
#ic2s2 I’ll be talking about this paper in one hour in Vingen 1+2!
Reposted by Anna Wegmann
#ic2s2 I’ll have a poster (#21) today on 𝐭𝐡𝐞 𝐢𝐦𝐩𝐚𝐜𝐭 𝐨𝐟 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐨𝐧 𝐬𝐨𝐜𝐢𝐚𝐥 𝐦𝐞𝐝𝐢𝐚 ✌🏼
What actually happens on social media when users get access to integrated AI?🤖
In our new preprint, we present a controlled experiment testing the impact of generative AI (4 treatments + control) on a social media platform, using a representative sample of nearly 700 U.S. participants.
In our new preprint, we present a controlled experiment testing the impact of generative AI (4 treatments + control) on a social media platform, using a representative sample of nearly 700 U.S. participants.
The Impact of Generative AI on Social Media: An Experimental Study
ai-research.andersgiovanni.com
July 22, 2025 at 8:12 AM
#ic2s2 I’ll have a poster (#21) today on 𝐭𝐡𝐞 𝐢𝐦𝐩𝐚𝐜𝐭 𝐨𝐟 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐨𝐧 𝐬𝐨𝐜𝐢𝐚𝐥 𝐦𝐞𝐝𝐢𝐚 ✌🏼