1 code implementation • 4 Mar 2024 • Kseniia Kuvshinova, Olga Tsymboi, Alina Kostromina, Dmitry Simakov, Elizaveta Kovtun
In this work, we consider the essential question if it is advantageous to train a foundation model on synthetic data or it is better to utilize only a limited number of real-life examples.
no code implementations • 20 Aug 2023 • Elizaveta Kovtun, Alina Ermilova, Dmitry Berestnev, Alexey Zaytsev
This adoption brings new danger, as deep black-box models are known to be vulnerable to adversarial attacks.
no code implementations • 24 Mar 2023 • Andrei Zaichenko, Aleksei Kazakov, Elizaveta Kovtun, Semen Budennyy
To start, we demonstrate the connection of tweet sentiments with the volatility of companies' stock prices.
1 code implementation • 1 Mar 2023 • Elizaveta Kovtun, Galina Boeva, Artem Zabolotnyi, Evgeny Burnaev, Martin Spindler, Alexey Zaytsev
For example, the micro-AUC of our approach is $0. 9536$ compared to $0. 7501$ for a vanilla transformer.
no code implementations • 18 Oct 2022 • Regina Tuganova, Anna Permyakova, Anna Kuznetsova, Karina Rakhmanova, Natalia Monzul, Roman Uvarov, Elizaveta Kovtun, Semen Budennyy
Our results show that the advancement of research in all three areas of eco-innovation does not automatically lead to the increase in the number of patents.
no code implementations • 11 Aug 2022 • Semen Budennyy, Alexey Kazakov, Elizaveta Kovtun, Leonid Zhukov
The first factor is the drug portfolio size of the company, indicating the greater susceptibility to an announcement in the case of small drug diversification.
1 code implementation • 15 Jun 2021 • Ivan Fursov, Matvey Morozov, Nina Kaploukhaya, Elizaveta Kovtun, Rodrigo Rivera-Castro, Gleb Gusev, Dmitry Babaev, Ivan Kireev, Alexey Zaytsev, Evgeny Burnaev
In this work, we examine adversarial attacks on transaction records data and defences from these attacks.