Donya Rooein (@donyarooein) header
Donya Rooein (@donyarooein) profile picture

AI for Education

Zurich
Joined July 2018
Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)Photo by Donya Rooein (@donyarooein)
Donya Rooein retweeted
A few snapshots from a brilliant week at @emnlpmeeting in Budapest! Lots of good conversations, old friends, and new connections!
6
1
26
1,180
Time flies and I am looking for full time research positions! I am finishing my PhD at Cambridge early 2027. My research focuses on simulating human behavior with LLMs, LLM personalization, evaluating and shaping model behavior, and understanding human–AI interaction. Please get in touch if you are hiring in relavant areas!
2
17
1
74
4,955
Donya Rooein retweeted
Hi @duolingo! Pittsburgh and CMU have given so much to help all kinds of people flourish, including your own non-US born co-founder and many others who helped make Pittsburgh and CMU the gems they are today. It breaks my heart to see a policy that discriminates based on place of birth. We don’t choose where we’re born, but we choose who we become. Please keep Duo true to its mission; let your tech serve all of humanity. 💚
5
96
3
480
12,857
Donya Rooein retweeted
PATS: Personality-Aware Teaching Strategies with LLM Tutors Aligning tutoring strategies with student personality improves learning interactions. 🗓 Wed, Mar 25 📍 Poster Session 3 (16:30–18:00) 🔗 aclanthology.org/2026.findin…
1
2
87
Donya Rooein retweeted
Can Reasoning Help LLMs Capture Human Annotator Disagreement? Across 60 setups, reasoning methods actually hurt disagreement modeling—especially in high-variance cases where humans genuinely disagree. 🗓 Fri, Mar 27 📍 Salle La Palmeraie 🔗 aclanthology.org/2026.eacl-l…
1
4
219
Donya Rooein retweeted
Exploring Subjective Tasks in Farsi (#WASSA) Farsi is under-resourced for subjective NLP tasks (sentiment, emotion, toxicity). The key issue isn’t just more data—but better, more representative data. 🗓 Sun, Mar 29 📍 Poster Hall 🔗 aclanthology.org/anthology-f…
1
2
143
Donya Rooein retweeted
Excited to share that @MilaNLProc will be presenting 5 new papers at #EACL2026 and workshops in Rabat 🇲🇦!
5
4
11
602
Donya Rooein retweeted
🎉 MilaNLP 2025 Wrapped 🎉 Lots of learning, building , sharing, and growing together 🌱 #NLProc
2
5
322
New Paper at #EACL2026 PATS: Personality-Aware Teaching Strategies with Large Language Model Tutors If you’re building LLM tutors, it's important to note whether they can adapt teaching strategies (not just answers) to support learners with different profiles.
1
3
142
🧵[3/4] We find that LLMs often provide high-quality QA with generic scaffolding or Motivational notes. But they rarely implement complex pedagogical moves tailored to the learner. However, if we instruct them with teaching strategies, they can implement some of them.
1
31
🧵[4/4] Teachers in our human studies preferred simulated conversations with LLM tutors that apply explicit teaching strategies. Big thanks to my co-authors and supervisors! #EACL2026 #NLP #LLMTutors #AIEducation #EdTech
52
New EACL paper!🥳 Reality check: humans disagree all the time 🤷‍♀️🤷‍♂️ In this paper, we ask: can LLMs help model that disagreement? 🧠⚖️📊
🚨 Using reasoning LLMs as annotators? You might be erasing critical human disagreements. Our EACL'25 paper shows RLVR-style reasoning actually HARMS disagreement modeling—even when carefully prompted to consider it! 🙀 📄 Paper: arxiv.org/abs/2506.19467🧵👇
3
152
Donya Rooein retweeted
“Teacher Demonstrations in a BabyLM’s Zone of Proximal Development for Contingent Multi-Turn Interaction” selected for an Outstanding Paper Award at the BabyLM Challenge & Workshop!
1
5
257
Donya Rooein retweeted
What an inspiring week at #EMNLP2025 in Suzhou🇨🇳! Huge thanks to the organizers and everyone who stopped by our poster/talk!
2
2
1
15
635
Donya Rooein retweeted
📖For our last Reading Group @donyarooein presented "Culture is Everywhere: A Call for Intentionally Cultural Evaluation" by Oh et al. Paper: arxiv.org/pdf/2509.01301 #NLProc
4
9
973
If you're at ACL, join us for the tutorial "LLMs for Education: Understanding the Needs of Stakeholders, Current Capabilities and the Path Forward" at the BEA workshop (Room 1.85–86) 9:00-12:30am tomorrow (July 31st) @aclmeeting
1
3
2
27
2,282
We have a great line of content bringing together ideas from the fields of NLP and AIED!
1
1
6
617
Contributions from: Sankalan Pal Chowdhury, Peng Cui, Nico Daheim, Ekaterina Kochmar, Jakub Macina, Yilmazcan Özyurt, Donya Rooein, Mrinmaya Sachan & Shashank Sonkar
1
5
459