Daryna's profile photo
Hi, I'm Daryna 👋🇺🇦 I am a postdoctoral researcher at Social Computing Research Group in Technical University of Munich🇩🇪. Before, I obtained my PhD degree at Skolkovo Institue of Science and Technology under supervision of Alexander Panchenko with topic "Method for Fighting Harmful Multilingual Textual Content" 📜. Currently, I continue to follow my research vector participating in eXplainable AI (XAI) project. More details in my CV.

Research

I am interested in applying Large Lange Models (in both monolingual and multilingual setups) to different task of NLP for Social good (NLP4SG). Moreover, I would like to make my solutions interpretable and efficient. The key topics I am currently focusing on are: Some of my works:

Multiverse illustration
Multiverse: Multilingual Evidence for Fake News Detection
D Dementieva, M Kuimov, A Panchenko
MDPI, 2023
Paper


Detoxification illustration
ParaDetox: Detoxification with Parallel Data
V Logacheva*, D Dementieva*, S Ustyantsev, D Moskovskiy, D Dale, I Krotova, N Semenov, A Panchenko (*equal contribution)
ACL, 2022
paper


Russian Detoxification illustration
Methods for detoxification of texts for the russian language
D Dementieva, D Moskovskiy, V Logacheva, D Dale, O Kozlova, N Semenov, A Panchenko
MDPI, 2021


IFAN illustration
IFAN: An Explainability-Focused Interaction Framework for Humans and NLP Models
E Mosca, D Dementieva, TE Ajdari, M Kummeth, K Gringauz, G Groh
In submission to ACL Demo 2023
paper

OpenSource Contribution🤗

Before, I contributed to s-nlp groud. More specifically,:

Now, I am one of the admins at tum-nlp.

Experience

Talks

Teaching

I am involved in both theoretical and practical NLP courses at TUM: Before, I had an amazing opportunity to participate in Statistical NLP and Neural NLP courses (Fall 2019, 2020, 2021) as TA at Skoltech.

Theses under my supervision:

Hackathons & Business Challenges

Contact

I am happy to get in touch! Please contact me at daryna.dementieva [at] tum.de or connect on LinkedIn.

This personal webpage was inspired by the example from amazing Anastasia Razdaibiedina.