FFE-Hallu:Hallucinations in Fixed Figurative Expressions:Benchmark of Idioms and Proverbs in the Persian Language
This addresses the problem of unreliable LLM outputs for users dealing with culturally grounded figurative language, particularly in underrepresented languages like Persian, though it is incremental as it focuses on benchmarking rather than solving the issue.
The researchers tackled the problem of figurative hallucination in large language models (LLMs) by creating FFEHallu, the first comprehensive benchmark for evaluating this issue in Persian, focusing on fixed figurative expressions like idioms and proverbs. They found that while some models like GPT4.1 performed well in certain tasks, most struggled to distinguish real expressions from fabrications and frequently hallucinated during translation, revealing systematic weaknesses in figurative competence.
Figurative language, particularly fixed figurative expressions (FFEs) such as idioms and proverbs, poses persistent challenges for large language models (LLMs). Unlike literal phrases, FFEs are culturally grounded, largely non-compositional, and conventionally fixed, making them especially vulnerable to figurative hallucination. We define figurative hallucination as the generation or endorsement of expressions that sound idiomatic and plausible but do not exist as authentic figurative expressions in the target language. We introduce FFEHallu, the first comprehensive benchmark for evaluating figurative hallucination in LLMs, with a focus on Persian, a linguistically rich yet underrepresented language. FFEHallu consists of 600 carefully curated instances spanning three complementary tasks: (i) FFE generation from meaning, (ii) detection of fabricated FFEs across four controlled construction categories, and (iii) FFE to FFE translation from English to Persian. Evaluating six state of the art multilingual LLMs, we find systematic weaknesses in figurative competence and cultural grounding. While models such as GPT4.1 demonstrate relatively strong performance in rejecting fabricated FFEs and retrieving authentic ones, most models struggle to reliably distinguish real expressions from high quality fabrications and frequently hallucinate during cross lingual translation. These findings reveal substantial gaps in current LLMs handling of figurative language and underscore the need for targeted benchmarks to assess and mitigate figurative hallucination.