Translating the Untranslatable: An Operationalizable Ontology for Untranslatability

2026-06-15

AI Summary

This study addresses the lack of systematic modeling and evaluation frameworks for handling untranslatability in machine translation. It introduces, for the first time, an operational ontology of untranslatability along with a taxonomy of compensation strategies, and constructs the first multilingual dataset annotated with such strategies—encompassing corpus collection, strategy annotation, and human preference evaluation. Experimental results demonstrate that translations employing annotation-based compensation strategies are significantly preferred by human evalu日消息ors, confirming the critical role of structured compensation approaches in enhancing translation quality. This work establishes both a theoretical foundation and a data resource for strategy-guided machine translation research.

Abstract

Untranslatability, cases where meaning cannot be directly preserved across languages, is well-studied in linguistics but underexplored in NLP. As machine translation (MT) systems improve on standard benchmarks, their limitations increasingly concentrate in such cases, where translation cannot be reduced to one-to-one equivalence. We introduce a structured ontology of untranslatability along with a taxonomy of compensation strategies, which are specific techniques to convey meaning under these untranslatable circumstances. We operationalize this framework into a multilingual dataset of untranslatable sentences paired with strategy-based translations, enabling controlled analysis of translation behavior. Initial human preference studies suggest that translation quality depends on the strategy used, with consistent preferences for outputs that include explanatory context, known as the Annotation compensation strategy. Our framework and dataset provide a foundation for studying and modeling strategy-informed machine translation.

Problem

• untranslatability
• machine translation
• compensation strategies
• NLP
• translation quality

Innovation

• untranslatability
• ontology
• compensation strategies
• machine translation
• multilingual dataset