Evolution of semantic similarity—a survey

D Chandrasekaran, V Mago - ACM Computing Surveys (CSUR), 2021 - dl.acm.org
Estimating the semantic similarity between text data is one of the challenging and open
research problems in the field of Natural Language Processing (NLP). The versatility of …

Analysis methods in neural language processing: A survey

Y Belinkov, J Glass - … of the Association for Computational Linguistics, 2019 - direct.mit.edu
The field of natural language processing has seen impressive progress in recent years, with
neural network models replacing many of the traditional systems. A plethora of new models …

How contextual are contextualized word representations? Comparing the geometry of BERT, ELMo, and GPT-2 embeddings

K Ethayarajh - arXiv preprint arXiv:1909.00512, 2019 - arxiv.org
Replacing static word embeddings with contextualized word representations has yielded
significant improvements on many NLP tasks. However, just how contextual are the …

Is bert really robust? a strong baseline for natural language attack on text classification and entailment

D Jin, Z Jin, JT Zhou, P Szolovits - Proceedings of the AAAI conference on …, 2020 - aaai.org
Abstract Machine learning algorithms are often vulnerable to adversarial examples that have
imperceptible alterations from the original counterparts but can fool the state-of-the-art …

Language models as knowledge bases?

F Petroni, T Rocktäschel, P Lewis, A Bakhtin… - arXiv preprint arXiv …, 2019 - arxiv.org
Recent progress in pretraining language models on large textual corpora led to a surge of
improvements for downstream NLP tasks. Whilst learning linguistic knowledge, these …

Probing pretrained language models for lexical semantics

I Vulić, EM Ponti, R Litschko, G Glavaš… - Proceedings of the …, 2020 - aclanthology.org
The success of large pretrained language models (LMs) such as BERT and RoBERTa has
sparked interest in probing their representations, in order to unveil what types of knowledge …

Null it out: Guarding protected attributes by iterative nullspace projection

S Ravfogel, Y Elazar, H Gonen, M Twiton… - arXiv preprint arXiv …, 2020 - arxiv.org
The ability to control for the kinds of information encoded in neural representation has a
variety of use cases, especially in light of the challenge of interpreting these models. We …

Senteval: An evaluation toolkit for universal sentence representations

A Conneau, D Kiela - arXiv preprint arXiv:1803.05449, 2018 - arxiv.org
We introduce SentEval, a toolkit for evaluating the quality of universal sentence
representations. SentEval encompasses a variety of tasks, including binary and multi-class …

WiC: the word-in-context dataset for evaluating context-sensitive meaning representations

MT Pilehvar, J Camacho-Collados - arXiv preprint arXiv:1808.09121, 2018 - arxiv.org
By design, word embeddings are unable to model the dynamic nature of words' semantics,
ie, the property of words to correspond to potentially different meanings. To address this …

Evaluating word embedding models: Methods and experimental results

B Wang, A Wang, F Chen, Y Wang… - APSIPA transactions on …, 2019 - cambridge.org
Extensive evaluation on a large number of word embedding models for language
processing applications is conducted in this work. First, we introduce popular word …