The rediscovery hypothesis: Language models need to meet linguistics V Nikoulina, M Tezekbayev, N Kozhakhmet, M Babazhanova, M Gallé, ... Journal of Artificial Intelligence Research 72, 1343-1384, 2021 | 5 | 2021 |
Speeding up entmax M Tezekbayev, V Nikoulina, M Gallé, Z Assylbekov arXiv preprint arXiv:2111.06832, 2021 | 2 | 2021 |
Geometric Probing of Word Vectors. M Babazhanova, M Tezekbayev, Z Assylbekov ESANN, 2021 | 2 | 2021 |
Autoencoders for a manifold learning problem with a jacobian rank constraint R Takhanov, YS Abylkairov, M Tezekbayev Pattern Recognition 143, 109777, 2023 | 1 | 2023 |
Intractability of Learning the Discrete Logarithm with Gradient-Based Methods R Takhanov, M Tezekbayev, A Pak, A Bolatov, Z Kadyrsizova, ... Asian Conference on Machine Learning, 1321-1336, 2024 | | 2024 |
Gradient Descent Fails to Learn High-frequency Functions and Modular Arithmetic R Takhanov, M Tezekbayev, A Pak, A Bolatov, Z Assylbekov arXiv preprint arXiv:2310.12660, 2023 | | 2023 |
Long-Tail Theory under Gaussian Mixtures A Bolatov, M Tezekbayev, I Melnykov, A Pak, V Nikoulina, Z Assylbekov arXiv preprint arXiv:2307.10736, 2023 | | 2023 |
Benign Overfitting with Retrieval Augmented Models Z Assylbekov, M Tezekbayev, V Nikoulina, M Gallé Nazarbayev University School of Sciences and Humanities, 2022 | | 2022 |
The rediscovery hypothesis: Language models need to meet linguistics Z Assylbekov, M Tezekbayev, N Kozhakhmet, V Nikoulina, M Gallé Journal of Artificial Intelligence Research, 2021 | | 2021 |
Semantics-and syntax-related subvectors in the skip-gram embeddings (student abstract) M Tezekbayev, Z Assylbekov, R Takhanov Proceedings of the AAAI Conference on Artificial Intelligence 34 (10), 13937 …, 2020 | | 2020 |
Semantics-and Syntax-related Subvectors in the Skip-gram Embeddings M Tezekbayev, Z Assylbekov, R Takhanov arXiv preprint arXiv:1912.13413, 2019 | | 2019 |