Dora Demszky

I'm an Assistant Professor in Education Data Science at the Stanford Graduate School of Education and in Computer Science (by courtesy). My lab's research combines machine learning, natural language processing, linguistics and input from practitioners, in order to develop interpretable and scalable education measures. We implement these measures as part of teacher-facing AI tools that go beyond productivity-enhancement to support teacher learning, for the purposes of advancing student learning.

I received my PhD in Linguistics at Stanford University, advised by Dan Jurafsky, and my BA in Linguistics with a minor in Computer Science summa cum laude at Princeton University.

🎓️

If you are interested in applying to do a PhD with me, please follow the application guidelines on the GSE website.

Unfortunately, I do not have the bandwidth to advise masters students outside of Stanford and high school students at this time.

🎉

News

Mei Tan's poster on meausing characteristics of feedback received the best paper award at the 2026 Stanford Education Data Science conference! 🏆
Paper led by Mei Tan, "Marked Pedagogies: Examining Linguistic Biases in Personalized Automated Writing Feedback", was a best paper runner-up at LAK '26 🏆 and featured in the Hechinger Report!
How Do Math Teachers Make Decisions About Using AI? : read our brief report from the 60-teacher summit we organized in June.
Our MathemaTikZ dataset received the inaugural best dataset prize at Learning at Scale! 🏆
Our NCTE classroom transcript dataset received the best IEDMS Publicly Available Educational Dataset Prize! 🏆
Our Practitioner Voices Summit in June hosted 60 teachers from 22 states, to provide them with a platform to inform R&D related to AI in classrooms. The rooms were buzzing with energy with all the ideas, and the dedication of these teachers to their students. We cannot wait to share what we learned!
How can we think about edtech in a socially responsible way? Read our take in this HAI blog-post my class co-authored! Here's also an accompanying piece in the GSE news.
The EduNLP website is live!
New School's In podcast episode with Dean Dan and Denise about how AI might support teachers!
My team was awarded a two-year grant from the Gates foundation to build tools that support teachers and multilingual students in using mathematical language practices in classrooms & scaffolding curricula to meet student needs.
Our team, lead by the amazing student Riz Malik won the 2024 Tools Competition with CoTeach.AI! The tool helps teachers scaffold math curricula to meet diverse learner needs.
In this short series put together by the Stanford Teacher Professional Learning team, listen to Bryan, Chris and I respond to educator questions about AI.
- How can AI support teacher training?
- How do you use AI to reduce teacher workload or improve processes?

Selected Publications

Do as I Say: What Teachers’ Language Reveals About Classroom Management Practices

Mei Tan, Dorottya Demszky

Educational Researcher, vol. 0, 2026, pp. 0013189X251410178

Paper

Mapping the Methodological Space of Classroom Interaction Research: Scale, Duration, and Modality in an Age of AI

Dorottya Demszky, Edith Bouton, Alison Twiner, Sara Hennessy, Richard Correnti

arXiv preprint arXiv:2604.28098, 2026

Paper

Marked Pedagogies: Examining Linguistic Biases in Personalized Automated Writing Feedback

Mei Tan, Lena Phalen, Dorottya Demszky

Proceedings of the 16th International Learning Analytics and Knowledge Conference (LAK '26), 2026

Paper

Mitigating LLM biases toward spurious social contexts using direct preference optimization

Hyunji Nam, Dorottya Demszky

arXiv preprint arXiv:2604.02585, 2026

Paper

Practitioner Voices Summit: How Teachers Evaluate AI Tools through Deliberative Sensemaking

Dorottya Demszky, Christopher Mah, Helen Higgins

arXiv preprint arXiv:2603.22588, 2026

Paper

EduCoder: An Open-Source Annotation System for Education Transcript Data

Saad Ashraf, James Malamut, Vishal Kumar, Guanzhong Pan, Mei Tan, Hyunji Nam, Lucia Langlois, Liliana Deonizio, Helen Higgins, Dorottya Demszky

ACL '26 System Demonstrations, 2026 Jul

Paper Try it! Github Demo video

IDEAlign: Comparing Large Language Models to Human Experts in Open-ended Interpretive Annotations

Hyunji Nam, Lucia Langlois, James Malamut, Mei Tan, Dorottya Demszky

Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics, Volume 1: Long Papers, 2026 Apr, pp. 3908–3925

Paper Github

TeachLM: Post-Training LLMs for Education Using Authentic Learning Data

Janos Perczel, Jin Chow, Dorottya Demszky

2025 Oct

Paper

Facilitating Evidence-Based Instructional Coaching With Automated Feedback on Teacher Discourse

James Malamut, Dorottya Demszky, Christine Bywater, Michele Reinhart, Heather C. Hill

EdWorkingPapers, 2025 Sep

Paper

MathemaTikZ: A Dataset and Benchmark for Mathematical Diagram Generation

Rizwaan Malik, Rebecca Li Hao, Ritika Kacholia, Dorottya Demszky

Proceedings of the Twelfth ACM Conference on Learning@ Scale, 2025 Jul, pp. 95--104

View PDF Dataset

Multi-Stage Speaker Diarization for Noisy Classrooms

Ali Sartaz Khan, Tolulope Ogunremi, Ahmed Adel Attia, Dorottya Demszky

Proceedings of the 18th International Conference on Educational Data Mining, 2025 Jul

View PDF

From Sentence-Corrections to Deeper Dialogue: Qualitative Insights from LLM and Teacher Feedback on Student Writing

Christopher Mah, Mei Tan, Lena Phalen, Alexa Sparks, Dorottya Demszky

EdWorkingPapers, 2025 Jun

Paper

Automated feedback improves teachers’ questioning quality in brick-and-mortar classrooms: Opportunities for further enhancement

Dorottya Demszky*, Jing Liu*, Heather C. Hill, Shyamoli Sanghi, Ariel Chung

Computers & Education, vol. 227, 2025 Apr

Paper

Does Increased Choice over Learning Topic Improve the Effectiveness of Automated Feedback for Educators?

Dorottya Demszky, Heather C Hill, Eric Taylor, Ashlee Kupor, Deepak Varuvel Dennison, Chris Piech

Education Sciences, vol. 15(9), 2025 Mar

Paper

Scaffolding Middle-School Mathematics Curricula With Large Language Models

Rizwaan Malik, Dorna Abdi, Rose Wang, Dorottya Demszky

British Journal of Education Technology, 2025 Jan

Paper CoTeach.AI HAI news

Computational Language Analysis Reveals that Process-Oriented Thinking About Belonging Aids the College Transition

Dorottya Demszky, C Lee Williams, Shannon T Brady, Shashanka Subrahmanya, Eric Gaudiello, Gregory M Walton, Johannes C Eichstaedt

2024

Paper

Problem-Oriented Segmentation and Retrieval: Case Study on Tutoring Conversations

Rose Wang, Pawan Wirawarn, Kenny Lam, Omar Khattab, Dorottya Demszky

Findings of the Association for Computational Linguistics: EMNLP 2024, 2024, pp. 12654--12672

Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertise

Rose E Wang, Ana T Ribeiro, Carly D Robinson, Susanna Loeb, Dorottya Demszky

arXiv preprint arXiv:2410.03017, 2024

Paper

An Open-Source Library for Education Conversation Data

Rose E Wang, Dorottya Demszky

NAACL System Demonstrations, 2024 Jun

Github Paper

Bridging the Novice-Expert Gap via Models of Decision-Making: A Case Study on Remediating Math Mistakes

Rose E. Wang, Qingyang Zhang, Carly Robinson, Susanna Loeb, Dorottya Demszky

Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), 2024 Jun

Paper Dan Meyer's blog post

Does Feedback on Talk Time Increase Student Engagement? Evidence from a Randomized Controlled Trial on a Math Tutoring Platform

Dorottya Demszky, Rose E Wang, Sean Geraghty, Carol Yu

The 14th Learning Analytics and Knowledge Conference (LAK '24), March 18--22, 2024, Kyoto, Japan, 2024 Mar

View PDF

“Mistakes Help Us Grow”: Facilitating and Evaluating Growth Mindset Supportive Language in Classrooms

Kunal Handa, Margaret Clapper, Jessica Boyle, Rose E Wang, Diyi Yang, David S Yeager, Dorottya Demszky

Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023 Oct

Paper

Using large language models in psychology

Dorottya Demszky*, Diyi Yang*, David S. Yeager*, Christopher J. Bryan, Margarett Clapper, Susannah Chandhok, Johannes C. Eichstaedt , Cameron Hecht, Jeremy Jamieson, Meghann Johnson, Michaela Jones, Danielle Krettek-Cobb, Leslie Lai, Nirel JonesMitchell, Desmond C. Ong, Carol S. Dweck, James J. Gross, James W. Pennebaker

Nature Reviews Psychology, 2023 Oct

M-Powering Teachers: Natural Language Processing Powered Feedback Improves 1:1 Instruction and Student Outcomes

Dorottya Demszky, Jing Liu

L@S '23: Proceedings of the Tenth ACM Conference on Learning @ Scale, 2023 Jul

View PDF Polygence

Is ChatGPT a Good Teacher Coach? Measuring Zero-Shot Performance For Scoring and Providing Actionable Insights on Classroom Instruction

Rose Wang, Dorottya Demszky

18th Workshop on Innovative Use of NLP for Building Educational Applications, 2023 Jun

View PDF

Can Automated Feedback Improve Teachers’ Uptake of Student Ideas? Evidence From a Randomized Controlled Trial In a Large-Scale Online Course

Dorottya Demszky, Jing Liu, Heather Hill, Dan Jurafsky, Chris Piech

Educational Evaluation and Policy Analysis, 2023 May

Paper

Content Analysis of Textbooks via Natural Language Processing: Findings on Gender, Race, and Ethnicity in Texas US History Textbooks

Li Lucy*, Dorottya Demszky*, Patricia Bromley, Dan Jurafsky

AERA Open, vol. 6, SAGE Publications Sage CA: Los Angeles, CA, 2020

View PDF Slides Code HAI Article

All Publications

Projects

M-Powering Teachers

Our research team and app that seeks to empower teachers by providing them with automated feedback.

In 2016, I co-founded a nonprofit organization, Tarisznya Alapítvány (Knapsack Foundation), the goal of which is to empower underprivileged children in Hungary through education. You can learn more on our website or on Facebook.

I am grateful to my incredible collaborators, mentors and mentees, who have made our projects possible and have made me a better person and researcher.