Papers by Maria Khodorchenko
SlideGuard: AI-Driven Evaluation of Graduate Student Presentation Materials (2026.acl-demo)
Copied to clipboard
Nikolay Alekseevich Butakov, Maria Khodorchenko, Mazein Nikita, Daniil Gareev, Yuri Falevskiy, Georgii Konev, Denis Nasonov
| Challenge: | Effective communication is a core objective of graduate education in AI and machine learning (ML). |
| Approach: | They propose an evaluation agent that assesses slide decks against a framework of expert-defined criteria using a visual language model. |
| Outcome: | The evaluation agent detects the majority of expert-identified issues on a dataset of 150 annotated slide decks and shows strong results on structural and visual criteria and known limitations on subjective dimensions such as research quality. |
Topic-driven Ensemble for Online Advertising Generation (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods for automating online advertising use open data . subdomains of text vary in use and can lead to reduced quality of adverts generation. |
| Approach: | They propose a neural network-based approach for the automatic generation of online advertising using texts from given webpages as sources. |
| Outcome: | The proposed approach significantly improves the quality of online advertising generated on a Russian dataset. |