FEED PETs: Further Experimentation and Expansion on the Disambiguation of Potentially Euphemistic Terms

Proceedings of the The 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023), pages 437–448 July 13-14, 2023 ©2023 Association for Computational Linguistics FEED PETs: Further Experimentation and Expansion on the Disambiguation of Potentially Euphemistic Terms Patrick Lee, Iyanuoluwa Shode , Alain Chirino Trujillo, Yuan Zhao, Olumide Ebenezer Ojo, Diana Cuevas Plancarte, Anna Feldman, Jing Peng Montclair State University New Jersey, USA {leep6,shodei1,chirinotruja1,zhaoy2,ojoo,cuevasplancd1,feldmana,pengj} @montclair.edu Abstract 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 ex- pand on the task in two ways. First, we an- notate PETs for vagueness, a linguistic prop- erty 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 Man- darin Chinese. We perform euphemism disam- biguation experiments in each language using multilingual transformer models mBERT and XLM-RoBERTa, establishing preliminary re- sults from which to launch future work. 1 Introduction Detecting and interpreting figurative language is a rapidly growing area in Natural Language Process- ing (NLP) (Chakrabarty et al., 2022; Liu and Hwa, 2017). Unfortunately, little work has been done on euphemism processing. Euphemisms are expres- sions that soften the message they convey. They are culture-specific and dynamic: they change over time. Therefore, dictionary-based approaches are ineffective (Bertram, 1998; Holder, 2002; Rawson, 2003). Euphemisms are often ambiguous: their figurative and non-figurative interpretation is of- ten context-dependent; see Table 1 for examples. Thus, existing work refers to these expressions as potentially euphemistic terms (PETs). State-of-the- art language models such as transformers perform well on many major NLP benchmarks. Recently, an attempt has been made to determine how these models perform in the euphemism disambiguation task (Lee et al., 2022a), in which an input text is classified as containing a euphemism or not. The described systems report promising results; how- ever, without further analysis and experimentation, it is unclear what transformers are capturing in or- der to perform the disambiguation, and the full extent of their ability in other languages. To address this, the present study describes two experiments to expand upon the euphemism disam- biguation task. In the first, we investigate a prag- matic property of euphemisms, vagueness, and use human annotations to distinguish between PETs which are more vague (vague euphemistic terms, or VETs) versus less vague. We then experiment with transformers’ abilities to disambiguate examples containing VETs versus non-VETs, and find that performance is generally higher for VETs. While we are unable to ascertain the exact reason for this discrepancy, we analyze the potential implications of the results and propose follow-up studies. In the second experiment, we create novel euphemism corpora for three other languages: Yorùbá, (Latin American and Castilian) Spanish, and Mandarin Chinese. Similarly to the English data, examples are obtained using a seed list of PETs, and include both euphemistic …

Content truncated to 500 words