FFE-Hallu:Hallucinations in Fixed Figurative Expressions:Benchmark of Idioms and Proverbs in the Persian Language

2026-01-27

AI Summary

This study addresses the phenomenon of “figurative hallucination” in large language models (LLMs)—the tendency to generate or misrecognize non-existent fixed figurative expressions, such as idioms—a problem particularly acute in low-resource languages like Persian. The work introduces the concept of figurative hallucination and presents FFE-Hallu, the first evaluation benchmark for Persian, comprising 600 expert-constructed samples across three tasks: paraphrase generation, fabrication detection, and English–Persian translation, with data quality ensured through controlled fabrication protocols. Systematic evaluation of prominent multilingual LLMs reveals pervasive deficiencies in cultural contextual understanding, with only GPT-4.1 demonstrating relative robustness on certain tasks. This research fills a critical gap in evaluating figurative language competence beyond English and advances the study of culturally embedded linguistic phenomena in LLMs.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Analogy

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

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.

Problem

• figurative hallucination
• fixed figurative expressions
• idioms
• proverbs
• large language models

Innovation

• figurative hallucination
• fixed figurative expressions
• benchmark
• Persian language
• multilingual LLMs