Challenges in Japanese Euphemism Classification: An Analysis of Pretrained Japanese and Multilingual Models

Noriko Takahashi, Whitney Poh, Libby Barak, JIng Peng, Anna Feldman


Abstract
Euphemisms present a persistent challenge for NLP because their interpretation depends on pragmatic inference, social norms, and contextual cues rather than surface meaning alone. Although Potentially Euphemistic Terms (PET)-based resources have been developed for several languages, Japanese euphemisms remain computationally unexplored despite their close interaction with honorifics, register variation, and orthographic choice. We introduce JP-PET, the first PET-based dataset for Japanese euphemism classification, comprising 1,672 annotated sentences across 101 PETs and ten semantic domains with register metadata. We evaluate two Japanese monolingual transformer models (Rinna RoBERTa and Tohoku BERT) and the multilingual XLM-R under three controlled PET-level data splits that isolate lexical familiarity and generalization to unseen euphemisms. While models achieve strong performance when PETs are shared between training and test data, performance drops substantially under PET-disjoint conditions, indicating reliance on lexical familiarity. Error analysis reveals systematic challenges in politically conventionalized expressions, metaphor-based euphemisms, and orthographic mitigation strategies. JP-PET provides the first benchmark for studying pragmatic meaning in Japanese NLP.
Anthology ID:
2026.nonliteral-1.1
Volume:
Proceedings of Learning Non-Literal Expressions with Small Data @ LREC 2026
Month:
May
Year:
2026
Address:
Palma, Mallorca (Spain)
Editors:
Markus Egg, Valia Kordoni
Venues:
NonLiteral | WS
SIG:
Publisher:
ELRA Language Resources Association (ELRA)
Note:
Pages:
1–11
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-nonliteral-01
DOI:
10.63317/234ssv2bkqd2
Bibkey:
Cite (ACL):
Noriko Takahashi, Whitney Poh, Libby Barak, JIng Peng, and Anna Feldman. 2026. Challenges in Japanese Euphemism Classification: An Analysis of Pretrained Japanese and Multilingual Models. In Proceedings of Learning Non-Literal Expressions with Small Data @ LREC 2026, pages 1–11, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
Challenges in Japanese Euphemism Classification: An Analysis of Pretrained Japanese and Multilingual Models (Takahashi et al., NonLiteral 2026)
Copy Citation: