Zobrazeno 1 - 6
of 6
pro vyhledávání: '"Cherniavskii, Daniil"'
Transformer-based language models have set new benchmarks across a wide range of NLP tasks, yet reliably estimating the uncertainty of their predictions remains a significant challenge. Existing uncertainty estimation (UE) techniques often fall short
Externí odkaz:
http://arxiv.org/abs/2308.11295
Autor:
Tulchinskii, Eduard, Kuznetsov, Kristian, Kushnareva, Laida, Cherniavskii, Daniil, Barannikov, Serguei, Piontkovskaya, Irina, Nikolenko, Sergey, Burnaev, Evgeny
Rapidly increasing quality of AI-generated content makes it difficult to distinguish between human and AI-generated texts, which may lead to undesirable consequences for society. Therefore, it becomes increasingly important to study the properties of
Externí odkaz:
http://arxiv.org/abs/2306.04723
Autor:
Trofimov, Ilya, Cherniavskii, Daniil, Tulchinskii, Eduard, Balabin, Nikita, Burnaev, Evgeny, Barannikov, Serguei
Publikováno v:
11th International Conference on Learning Representations (ICLR 2023)
We propose a method for learning topology-preserving data representations (dimensionality reduction). The method aims to provide topological similarity between the data manifold and its latent representation via enforcing the similarity in topologica
Externí odkaz:
http://arxiv.org/abs/2302.00136
Autor:
Tulchinskii, Eduard, Kuznetsov, Kristian, Kushnareva, Laida, Cherniavskii, Daniil, Barannikov, Serguei, Piontkovskaya, Irina, Nikolenko, Sergey, Burnaev, Evgeny
Publikováno v:
Proc. INTERSPEECH 2023, pages 311--315
We apply topological data analysis (TDA) to speech classification problems and to the introspection of a pretrained speech model, HuBERT. To this end, we introduce a number of topological and algebraic features derived from Transformer attention maps
Externí odkaz:
http://arxiv.org/abs/2211.17223
Autor:
Cherniavskii, Daniil, Tulchinskii, Eduard, Mikhailov, Vladislav, Proskurina, Irina, Kushnareva, Laida, Artemova, Ekaterina, Barannikov, Serguei, Piontkovskaya, Irina, Piontkovski, Dmitri, Burnaev, Evgeny
Publikováno v:
Findings of the Association for Computational Linguistics: EMNLP 2022, 88-107
The role of the attention mechanism in encoding linguistic knowledge has received special interest in NLP. However, the ability of the attention heads to judge the grammatical acceptability of a sentence has been underexplored. This paper approaches
Externí odkaz:
http://arxiv.org/abs/2205.09630
Autor:
Kushnareva, Laida, Cherniavskii, Daniil, Mikhailov, Vladislav, Artemova, Ekaterina, Barannikov, Serguei, Bernstein, Alexander, Piontkovskaya, Irina, Piontkovski, Dmitri, Burnaev, Evgeny
Publikováno v:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 635-649
The impressive capabilities of recent generative models to create texts that are challenging to distinguish from the human-written ones can be misused for generating fake news, product reviews, and even abusive content. Despite the prominent performa
Externí odkaz:
http://arxiv.org/abs/2109.04825