[HTML][HTML] Extracting sentence embeddings from pretrained transformer models

L Stankevičius, M Lukoševičius - Applied Sciences, 2024 - mdpi.com
Pre-trained transformer models shine in many natural language processing tasks and
therefore are expected to bear the representation of the input sentence or text meaning …

Context-aware cross-lingual map**

H Aldarmaki, M Diab - arxiv preprint arxiv:1903.03243, 2019 - arxiv.org
Cross-lingual word vectors are typically obtained by fitting an orthogonal matrix that maps
the entries of a bilingual dictionary from a source to a target vector space. Word vectors …

Efficient sentence embedding using discrete cosine transform

N Almarwani, H Aldarmaki, M Diab - arxiv preprint arxiv:1909.03104, 2019 - arxiv.org
Vector averaging remains one of the most popular sentence embedding methods in spite of
its obvious disregard for syntactic structure. While more complex sequential or convolutional …

A qualitative evaluation framework for paraphrase identification

V Kovatchev, MA Marti, M Salamo… - Proceedings of the …, 2019 - aclanthology.org
In this paper, we present a new approach for the evaluation, error analysis, and
interpretation of supervised and unsupervised Paraphrase Identification (PI) systems. Our …

Unsupervised sentence-embeddings by manifold approximation and projection

S Kayal - arxiv preprint arxiv:2102.03795, 2021 - arxiv.org
The concept of unsupervised universal sentence encoders has gained traction recently,
wherein pre-trained models generate effective task-agnostic fixed-dimensional …

Decomposing and comparing meaning relations: Paraphrasing, textual entailment, contradiction, and specificity

V Kovatchev, D Gold, MA Marti… - Proceedings of the …, 2020 - aclanthology.org
In this paper, we present a methodology for decomposing and comparing multiple meaning
relations (paraphrasing, textual entailment, contradiction, and specificity). The methodology …

Compressing Sentence Representation with Maximum Coding Rate Reduction

D Ševerdija, T Prusina, A Jovanović… - 2023 46th MIPRO …, 2023 - ieeexplore.ieee.org
In most natural language inference problems, sentence representation is needed for
semantic retrieval tasks. In recent years, pre-trained large language models have been quite …

Scalable cross-lingual transfer of neural sentence embeddings

H Aldarmaki, M Diab - arxiv preprint arxiv:1904.05542, 2019 - arxiv.org
We develop and investigate several cross-lingual alignment approaches for neural sentence
embedding models, such as the supervised inference classifier, InferSent, and sequential …

Cross-Lingual Alignment of Word & Sentence Embeddings

H Aldarmaki - 2019 - search.proquest.com
One of the notable developments in current natural language processing is the practical
efficacy of probabilistic word representations, where words are embedded in high …

[BOK][B] Automatic Summarization of Financial Reports

M Alikhani - 2021 - search.proquest.com
Abstract The field of Natural Language Processing (NLP) has witnessed substantial
advancements due to both the development of new algorithms and the increase of …