Anna Feldman

Iyanuoluwa Shode, David Ifeoluwa Adelani, Anna Feldman

A movie that is thoroughly enjoyed and recommended by an individual might be hated by another. One characteristic of humans is the ability to have feelings which could be positive or negative. To automatically classify and study human feelings, an aspect of natural language processing, sentiment analysis and opinion mining were designed to understand human feelings regarding several issues which could affect a product, a social media platforms, government, or societal discussions or even movies. Several works on sentiment analysis have been done on high resource languages while low resources languages like Yoruba have been sidelined. Due to the scarcity of datasets and linguistic architectures that will suit low resource languages, African languages "low resource languages" have been ignored and not fully explored. For this reason, our attention is placed on Yoruba to explore sentiment analysis on reviews of Nigerian movies. The data comprised 1500 movie reviews that were sourced from IMDB, Rotten Tomatoes, Letterboxd, Cinemapointer and Nollyrated. We develop sentiment classification models using the state-of-the-art pre-trained language models like mBERT and AfriBERTa to classify the movie reviews.

Martha Gavidia, Patrick Lee, Anna Feldman et al.

Euphemisms have not received much attention in natural language processing, despite being an important element of polite and figurative language. Euphemisms prove to be a difficult topic, not only because they are subject to language change, but also because humans may not agree on what is a euphemism and what is not. Nevertheless, the first step to tackling the issue is to collect and analyze examples of euphemisms. We present a corpus of potentially euphemistic terms (PETs) along with example texts from the GloWbE corpus. Additionally, we present a subcorpus of texts where these PETs are not being used euphemistically, which may be useful for future applications. We also discuss the results of multiple analyses run on the corpus. Firstly, we find that sentiment analysis on the euphemistic texts supports that PETs generally decrease negative and offensive sentiment. Secondly, we observe cases of disagreement in an annotation task, where humans are asked to label PETs as euphemistic or not in a subset of our corpus text examples. We attribute the disagreement to a variety of potential reasons, including if the PET was a commonly accepted term (CAT).

Patrick Lee, Martha Gavidia, Anna Feldman et al.

This paper presents a linguistically driven proof of concept for finding potentially euphemistic terms, or PETs. Acknowledging that PETs tend to be commonly used expressions for a certain range of sensitive topics, we make use of distributional similarities to select and filter phrase candidates from a sentence and rank them using a set of simple sentiment-based metrics. We present the results of our approach tested on a corpus of sentences containing euphemisms, demonstrating its efficacy for detecting single and multi-word PETs from a broad range of topics. We also discuss future potential for sentiment-based methods on this task.

Patrick Lee, Anna Feldman, Jing Peng

This paper presents The Shared Task on Euphemism Detection for the Third Workshop on Figurative Language Processing (FigLang 2022) held in conjunction with EMNLP 2022. Participants were invited to investigate the euphemism detection task: given input text, identify whether it contains a euphemism. The input data is a corpus of sentences containing potentially euphemistic terms (PETs) collected from the GloWbE corpus (Davies and Fuchs, 2015), and are human-annotated as containing either a euphemistic or literal usage of a PET. In this paper, we present the results and analyze the common themes, methods and findings of the participating teams

Olumide E. Ojo, Olaronke O. Adebanji, Hiram Calvo et al.

In this paper, we share our best performing submission to the Arabic AI Tasks Evaluation Challenge (ArAIEval) at ArabicNLP 2023. Our focus was on Task 1, which involves identifying persuasion techniques in excerpts from tweets and news articles. The persuasion technique in Arabic texts was detected using a training loop with XLM-RoBERTa, a language-agnostic text representation model. This approach proved to be potent, leveraging fine-tuning of a multilingual language model. In our evaluation of the test set, we achieved a micro F1 score of 0.64 for subtask A of the competition.

Olumide E. Ojo, Olaronke O. Adebanji, Alexander Gelbukh et al.

Zero-shot classification enables text to be classified into classes not seen during training. In this study, we examine the efficacy of zero-shot learning models in classifying healthcare consultation responses from Doctors and AI systems. The models evaluated include BART, BERT, XLM, XLM-R and DistilBERT. The models were tested on three different datasets based on a binary and multi-label analysis to identify the origins of text in health consultations without any prior corpus training. According to our findings, the zero-shot language models show a good understanding of language generally, but has limitations when trying to classify doctor and AI responses to healthcare consultations. This research provides a foundation for future research in the field of medical text classification by informing the development of more accurate methods of classifying text written by Doctors and AI systems in health consultations.

Hasan Can Biyik, Libby Barak, Jing Peng et al.

Euphemisms substitute socially sensitive expressions, often softening or reframing meaning, and their reliance on cultural and pragmatic context complicates modeling across languages. In this study, we investigate how cross-lingual equivalence influences transfer in multilingual euphemism detection. We categorize Potentially Euphemistic Terms (PETs) in Turkish and English into Overlapping (OPETs) and Non-Overlapping (NOPETs) subsets based on their functional, pragmatic, and semantic alignment. Our findings reveal a transfer asymmetry: semantic overlap is insufficient to guarantee positive transfer, particularly in low-resource Turkish-to-English direction, where performance can degrade even for overlapping euphemisms, and in some cases, improve under NOPET-based training. Differences in label distribution help explain these counterintuitive results. Category-level analysis suggests that transfer may be influenced by domain-specific alignment, though evidence is limited by sparsity.

Hasan Can Biyik, Patrick Lee, Anna Feldman

Euphemisms are a form of figurative language relatively understudied in natural language processing. This research extends the current computational work on potentially euphemistic terms (PETs) to Turkish. We introduce the Turkish PET dataset, the first available of its kind in the field. By creating a list of euphemisms in Turkish, collecting example contexts, and annotating them, we provide both euphemistic and non-euphemistic examples of PETs in Turkish. We describe the dataset and methodologies, and also experiment with transformer-based models on Turkish euphemism detection by using our dataset for binary classification. We compare performances across models using F1, accuracy, and precision as evaluation metrics.

Olumide Ebenezer Ojo, Olaronke Oluwayemisi Adebanji, Alexander Gelbukh et al.

Effective communication between healthcare providers and patients is crucial to providing high-quality patient care. In this work, we investigate how Doctor-written and AI-generated texts in healthcare consultations can be classified using state-of-the-art embeddings and one-shot classification systems. By analyzing embeddings such as bag-of-words, character n-grams, Word2Vec, GloVe, fastText, and GPT2 embeddings, we examine how well our one-shot classification systems capture semantic information within medical consultations. Results show that the embeddings are capable of capturing semantic features from text in a reliable and adaptable manner. Overall, Word2Vec, GloVe and Character n-grams embeddings performed well, indicating their suitability for modeling targeted to this task. GPT2 embedding also shows notable performance, indicating its suitability for models tailored to this task as well. Our machine learning architectures significantly improved the quality of health conversations when training data are scarce, improving communication between patients and healthcare providers.

Julia Sammartino, Libby Barak, Jing Peng et al.

Euphemisms are culturally variable and often ambiguous, posing challenges for language models, especially in low-resource settings. This paper investigates how cross-lingual transfer via sequential fine-tuning affects euphemism detection across five languages: English, Spanish, Chinese, Turkish, and Yoruba. We compare sequential fine-tuning with monolingual and simultaneous fine-tuning using XLM-R and mBERT, analyzing how performance is shaped by language pairings, typological features, and pretraining coverage. Results show that sequential fine-tuning with a high-resource L1 improves L2 performance, especially for low-resource languages like Yoruba and Turkish. XLM-R achieves larger gains but is more sensitive to pretraining gaps and catastrophic forgetting, while mBERT yields more stable, though lower, results. These findings highlight sequential fine-tuning as a simple yet effective strategy for improving euphemism detection in multilingual models, particularly when low-resource languages are involved.

Michael Flor, Xinyi Liu, Anna Feldman

Idioms are figurative expressions whose meanings often cannot be inferred from their individual words, making them difficult to process computationally and posing challenges for human experimental studies. This survey reviews datasets developed in psycholinguistics and computational linguistics for studying idioms, focusing on their content, form, and intended use. Psycholinguistic resources typically contain normed ratings along dimensions such as familiarity, transparency, and compositionality, while computational datasets support tasks like idiomaticity detection/classification, paraphrasing, and cross-lingual modeling. We present trends in annotation practices, coverage, and task framing across 53 datasets. Although recent efforts expanded language coverage and task diversity, there seems to be no relation yet between psycholinguistic and computational research on idioms.

Patrick Lee, Alain Chirino Trujillo, Diana Cuevas Plancarte et al.

This study investigates the computational processing of euphemisms, a universal linguistic phenomenon, across multiple languages. We train a multilingual transformer model (XLM-RoBERTa) to disambiguate potentially euphemistic terms (PETs) in multilingual and cross-lingual settings. In line with current trends, we demonstrate that zero-shot learning across languages takes place. We also show cases where multilingual models perform better on the task compared to monolingual models by a statistically significant margin, indicating that multilingual data presents additional opportunities for models to learn about cross-lingual, computational properties of euphemisms. In a follow-up analysis, we focus on universal euphemistic "categories" such as death and bodily functions among others. We test to see whether cross-lingual data of the same domain is more important than within-language data of other domains to further understand the nature of the cross-lingual transfer.

Patrick Lee, Iyanuoluwa Shode, Alain Chirino Trujillo et al.

Transformers have been shown to work well for the task of English euphemism disambiguation, in which a potentially euphemistic term (PET) is classified as euphemistic or non-euphemistic in a particular context. In this study, we expand on the task in two ways. First, we annotate PETs for vagueness, a linguistic property associated with euphemisms, and find that transformers are generally better at classifying vague PETs, suggesting linguistic differences in the data that impact performance. Second, we present novel euphemism corpora in three different languages: Yoruba, Spanish, and Mandarin Chinese. We perform euphemism disambiguation experiments in each language using multilingual transformer models mBERT and XLM-RoBERTa, establishing preliminary results from which to launch future work.

Iyanuoluwa Shode, David Ifeoluwa Adelani, Jing Peng et al.

Africa has over 2000 indigenous languages but they are under-represented in NLP research due to lack of datasets. In recent years, there have been progress in developing labeled corpora for African languages. However, they are often available in a single domain and may not generalize to other domains. In this paper, we focus on the task of sentiment classification for cross domain adaptation. We create a new dataset, NollySenti - based on the Nollywood movie reviews for five languages widely spoken in Nigeria (English, Hausa, Igbo, Nigerian-Pidgin, and Yoruba. We provide an extensive empirical evaluation using classical machine learning methods and pre-trained language models. Leveraging transfer learning, we compare the performance of cross-domain adaptation from Twitter domain, and cross-lingual adaptation from English language. Our evaluation shows that transfer from English in the same target domain leads to more than 5% improvement in accuracy compared to transfer from Twitter in the same language. To further mitigate the domain difference, we leverage machine translation (MT) from English to other Nigerian languages, which leads to a further improvement of 7% over cross-lingual evaluation. While MT to low-resource languages are often of low quality, through human evaluation, we show that most of the translated sentences preserve the sentiment of the original English reviews.

Shaden Shaar, Firoj Alam, Giovanni Da San Martino et al.

We present the results and the main findings of the NLP4IF-2021 shared tasks. Task 1 focused on fighting the COVID-19 infodemic in social media, and it was offered in Arabic, Bulgarian, and English. Given a tweet, it asked to predict whether that tweet contains a verifiable claim, and if so, whether it is likely to be false, is of general interest, is likely to be harmful, and is worthy of manual fact-checking; also, whether it is harmful to society, and whether it requires the attention of policy makers. Task~2 focused on censorship detection, and was offered in Chinese. A total of ten teams submitted systems for task 1, and one team participated in task 2; nine teams also submitted a system description paper. Here, we present the tasks, analyze the results, and discuss the system submissions and the methods they used. Most submissions achieved sizable improvements over several baselines, and the best systems used pre-trained Transformers and ensembles. The data, the scorers and the leaderboards for the tasks are available at http://gitlab.com/NLP4IF/nlp4if-2021.

Kei Yin Ng, Anna Feldman, Jing Peng

This paper studies how the linguistic components of blogposts collected from Sina Weibo, a Chinese microblogging platform, might affect the blogposts' likelihood of being censored. Our results go along with King et al. (2013)'s Collective Action Potential (CAP) theory, which states that a blogpost's potential of causing riot or assembly in real life is the key determinant of it getting censored. Although there is not a definitive measure of this construct, the linguistic features that we identify as discriminatory go along with the CAP theory. We build a classifier that significantly outperforms non-expert humans in predicting whether a blogpost will be censored. The crowdsourcing results suggest that while humans tend to see censored blogposts as more controversial and more likely to trigger action in real life than the uncensored counterparts, they in general cannot make a better guess than our model when it comes to `reading the mind' of the censors in deciding whether a blogpost should be censored. We do not claim that censorship is only determined by the linguistic features. There are many other factors contributing to censorship decisions. The focus of the present paper is on the linguistic form of blogposts. Our work suggests that it is possible to use linguistic properties of social media posts to automatically predict if they are going to be censored.

Kei Yin Ng, Anna Feldman, Jing Peng et al.

This paper investigates censorship from a linguistic perspective. We collect a corpus of censored and uncensored posts on a number of topics, build a classifier that predicts censorship decisions independent of discussion topics. Our investigation reveals that the strongest linguistic indicator of censored content of our corpus is its readability.

Jing Peng, Anna Feldman, Ekaterina Vylomova

We describe an algorithm for automatic classification of idiomatic and literal expressions. Our starting point is that words in a given text segment, such as a paragraph, that are highranking representatives of a common topic of discussion are less likely to be a part of an idiomatic expression. Our additional hypothesis is that contexts in which idioms occur, typically, are more affective and therefore, we incorporate a simple analysis of the intensity of the emotions expressed by the contexts. We investigate the bag of words topic representation of one to three paragraphs containing an expression that should be classified as idiomatic or literal (a target phrase). We extract topics from paragraphs containing idioms and from paragraphs containing literals using an unsupervised clustering method, Latent Dirichlet Allocation (LDA) (Blei et al., 2003). Since idiomatic expressions exhibit the property of non-compositionality, we assume that they usually present different semantics than the words used in the local topic. We treat idioms as semantic outliers, and the identification of a semantic shift as outlier detection. Thus, this topic representation allows us to differentiate idioms from literals using local semantic contexts. Our results are encouraging.