Search Results for author: Dmitry Nikolaev

Found 24 papers, 7 papers with code

Adverbs, Surprisingly

no code implementations31 May 2023 Dmitry Nikolaev, Collin F. Baker, Miriam R. L. Petruck, Sebastian Padó

This paper begins with the premise that adverbs are neglected in computational linguistics.

Language Modelling

Additive manifesto decomposition: A policy domain aware method for understanding party positioning

1 code implementation17 May 2023 Tanise Ceron, Dmitry Nikolaev, Sebastian Padó

The workflow covers (a) definition of suitable policy domains; (b) automatic labeling of domains, if no manual labels are available; (c) computation of domain-level similarities and aggregation at a global level; (d) extraction of interpretable party positions on major policy axes via multidimensional scaling.

Representation biases in sentence transformers

no code implementations30 Jan 2023 Dmitry Nikolaev, Sebastian Padó

Variants of the BERT architecture specialised for producing full-sentence representations often achieve better performance on downstream tasks than sentence embeddings extracted from vanilla BERT.

Sentence Embeddings

Word-order typology in Multilingual BERT: A case study in subordinate-clause detection

no code implementations NAACL (SIGTYP) 2022 Dmitry Nikolaev, Sebastian Padó

The capabilities and limitations of BERT and similar models are still unclear when it comes to learning syntactic abstractions, in particular across languages.

Fast matrix multiplication for binary and ternary CNNs on ARM CPU

no code implementations18 May 2022 Anton Trusov, Elena Limonova, Dmitry Nikolaev, Vladimir V. Arlazarov

In this paper, we propose novel fast algorithms of ternary, ternary-binary, and binary matrix multiplication for mobile devices with ARM architecture.

On the properties of some low-parameter models for color reproduction in terms of spectrum transformations and coverage of a color triangle

no code implementations21 Oct 2021 Alexey Kroshnin, Viacheslav Vasilev, Egor Ershov, Denis Shepelev, Dmitry Nikolaev, Mikhail Tchobanou

One of the classical approaches to solving color reproduction problems, such as color adaptation or color space transform, is the use of low-parameter spectral models.

On the Relation between Syntactic Divergence and Zero-Shot Performance

1 code implementation EMNLP 2021 Ofir Arviv, Dmitry Nikolaev, Taelin Karidi, Omri Abend

We explore the link between the extent to which syntactic relations are preserved in translation and the ease of correctly constructing a parse tree in a zero-shot setting.

Cross-lingual zero-shot dependency parsing Relation Classification

Part of Speech and Universal Dependency effects on English Arabic Machine Translation

no code implementations1 Jun 2021 Ofek Rafaeli, Omri Abend, Leshem Choshen, Dmitry Nikolaev

In this research paper, I will elaborate on a method to evaluate machine translation models based on their performance on underlying syntactical phenomena between English and Arabic languages.

BIG-bench Machine Learning Machine Translation +1

SERRANT: a syntactic classifier for English Grammatical Error Types

1 code implementation6 Apr 2021 Leshem Choshen, Matanel Oren, Dmitry Nikolaev, Omri Abend

SERRANT is a system and code for automatic classification of English grammatical errors that combines SErCl and ERRANT.

General Classification

Illumination Estimation Challenge: experience of past two years

no code implementations31 Dec 2020 Egor Ershov, Alex Savchik, Ilya Semenkov, Nikola Banić, Karlo Koscević, Marko Subašić, Alexander Belokopytov, Zhihao LI, Arseniy Terekhin, Daria Senshina, Artem Nikonorov, Yanlin Qian, Marco Buzzelli, Riccardo Riva, Simone Bianco, Raimondo Schettini, Sven Lončarić, Dmitry Nikolaev

The main advantage of testing a method on a challenge over testing in on some of the known datasets is the fact that the ground-truth illuminations for the challenge test images are unknown up until the results have been submitted, which prevents any potential hyperparameter tuning that may be biased.

Color Constancy Vocal Bursts Valence Prediction

Classifying Syntactic Errors in Learner Language

1 code implementation CONLL 2020 Leshem Choshen, Dmitry Nikolaev, Yevgeni Berzak, Omri Abend

We present a method for classifying syntactic errors in learner language, namely errors whose correction alters the morphosyntactic structure of a sentence.

Classification General Classification +1

ResNet-like Architecture with Low Hardware Requirements

1 code implementation15 Sep 2020 Elena Limonova, Daniil Alfonso, Dmitry Nikolaev, Vladimir V. Arlazarov

In the paper, we introduce a bipolar morphological ResNet (BM-ResNet) model obtained from a much more complex ResNet architecture by converting its layers to bipolar morphological ones.

Edge-computing General Classification +1

Fast Implementation of 4-bit Convolutional Neural Networks for Mobile Devices

no code implementations14 Sep 2020 Anton Trusov, Elena Limonova, Dmitry Slugin, Dmitry Nikolaev, Vladimir V. Arlazarov

We introduce an efficient implementation of 4-bit matrix multiplication for quantized neural networks and perform time measurements on a mobile ARM processor.

Optical Character Recognition (OCR) Quantization

Line detection via a lightweight CNN with a Hough Layer

no code implementations20 Aug 2020 Lev Teplyakov, Kirill Kaymakov, Evgeny Shvets, Dmitry Nikolaev

Line detection is an important computer vision task traditionally solved by Hough Transform.

Line Detection

Accelerated FBP for computed tomography image reconstruction

no code implementations13 Jul 2020 Anastasiya Dolmatova, Marina Chukalina, Dmitry Nikolaev

The classical direct implementations of this algorithm require the execution of $\Theta(N^3)$ operations, where $N$ is the linear size of the 2D slice.

Image Reconstruction

Fine-Grained Analysis of Cross-Linguistic Syntactic Divergences

1 code implementation ACL 2020 Dmitry Nikolaev, Ofir Arviv, Taelin Karidi, Neta Kenneth, Veronika Mitnik, Lilja Maria Saeboe, Omri Abend

The patterns in which the syntax of different languages converges and diverges are often used to inform work on cross-lingual transfer.

Cross-Lingual Transfer

SegBo: A Database of Borrowed Sounds in the World's Languages

no code implementations LREC 2020 Eitan Grossman, Elad Eisen, Dmitry Nikolaev, Steven Moran

Phonological segment borrowing is a process through which languages acquire new contrastive speech sounds as the result of borrowing new words from other languages.

Fast Implementation of Morphological Filtering Using ARM NEON Extension

no code implementations19 Feb 2020 Elena Limonova, Arseny Terekhin, Dmitry Nikolaev, Vladimir Arlazarov

Experiments showed 3 times efficiency increase for final implementation of erosion and dilation compared to van Herk/Gil-Werman algorithm without SIMD, 5. 7 times speedup for 8x8 matrix transpose and 12 times speedup for 16x16 matrix transpose compared to transpose without SIMD.

Computational optimization of convolutional neural networks using separated filters architecture

no code implementations18 Feb 2020 Elena Limonova, Alexander Sheshkus, Dmitry Nikolaev

This paper considers a convolutional neural network transformation that reduces computation complexity and thus speedups neural network processing.

A Document Skew Detection Method Using Fast Hough Transform

no code implementations5 Dec 2019 Pavel Bezmaternykh, Dmitry Nikolaev

The majority of document image analysis systems use a document skew detection algorithm to simplify all its further processing stages.

A Method of Detecting End-To-End Curves of Limited Curvature

no code implementations4 Dec 2019 Ekaterina Panfilova, Mikhail Aliev, Irina Kunina, Vasiliy Postnikov, Dmitry Nikolaev

In this paper we consider a method for detecting end-to-end curves of limited curvature like the k-link polylines with bending angle between adjacent segments in a given range.

Bipolar Morphological Neural Networks: Convolution Without Multiplication

no code implementations5 Nov 2019 Elena Limonova, Daniil Matveev, Dmitry Nikolaev, Vladimir V. Arlazarov

To demonstrate efficiency of the proposed model we consider classical convolutional neural networks and convert the pre-trained convolutional layers to the bipolar morphological layers.

Linear colour segmentation revisited

2 code implementations2 Jan 2019 Anna Smagina, Valentina Bozhkova, Sergey Gladilin, Dmitry Nikolaev

In this work we discuss the known algorithms for linear colour segmentation based on a physical approach and propose a new modification of segmentation algorithm.

Cannot find the paper you are looking for? You can Submit a new open access paper.