# PerHalluEval: Persian Hallucination Evaluation Benchmark for Large Language Models

arXiv: https://arxiv.org/abs/2509.21104 · HTML: https://mlsanity.com/paper/2509.21104/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 47 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Builds on (most cited first, 24 of 24)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [On Faithfulness and Factuality in Abstractive Summarization](https://mlsanity.com/paper/2005.00661/) (2020)
- [SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models](https://mlsanity.com/paper/2303.08896/) (2023)
- [Evaluating the Factual Consistency of Abstractive Text Summarization](https://mlsanity.com/paper/1910.12840/) (2019)
- [Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond](https://mlsanity.com/paper/2304.13712/) (2023)
- [Asking and Answering Questions to Evaluate the Factual Consistency of Summaries](https://mlsanity.com/paper/2004.04228/) (2020)
- [HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models](https://mlsanity.com/paper/2305.11747/) (2023)
- [FEQA: A Question Answering Evaluation Framework for Faithfulness Assessment in Abstractive Summarization](https://mlsanity.com/paper/2005.03754/) (2020)
- [Evaluating Attribution in Dialogue Systems: The BEGIN Benchmark](https://mlsanity.com/paper/2105.00071/) (2021)
- [Multi-Fact Correction in Abstractive Text Summarization](https://mlsanity.com/paper/2010.02443/) (2020)
- [ParsiNLU: A Suite of Language Understanding Challenges for Persian](https://mlsanity.com/paper/2012.06154/) (2020)
- [FactCHD: Benchmarking Fact-Conflicting Hallucination Detection](https://mlsanity.com/paper/2310.12086/) (2023)
- [HalluDial: A Large-Scale Benchmark for Automatic Dialogue-Level Hallucination Evaluation](https://mlsanity.com/paper/2406.07070/) (2024)
- [FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluation](https://mlsanity.com/paper/2410.22257/) (2024)
- [Log Probabilities Are a Reliable Estimate of Semantic Plausibility in Base and Instruction-Tuned Language Models](https://mlsanity.com/paper/2403.14859/) (2024)
- [Khayyam Challenge (PersianMMLU): Is Your LLM Truly Wise to The Persian Language?](https://mlsanity.com/paper/2404.06644/) (2024)
- [HELP ME THINK: A Simple Prompting Strategy for Non-experts to Create Customized Content with Models](https://mlsanity.com/paper/2208.08232/) (2022)
- [PQuAD: A Persian Question Answering Dataset](https://mlsanity.com/paper/2202.06219/) (2022)
- [Poly-FEVER: A Multilingual Fact Verification Benchmark for Hallucination Detection in Large Language Models](https://mlsanity.com/paper/2503.16541/) (2025)
- [naab: A ready-to-use plug-and-play corpus for Farsi](https://mlsanity.com/paper/2208.13486/) (2022)
- [evaluating bert and parsbert for analyzing persian advertisement data](https://mlsanity.com/paper/2305.02426/) (2023)
- [NeuroBlend: Towards Low-Power yet Accurate Neural Network-Based Inference Engine Blending Binary and Fixed-Point Convolutions](https://mlsanity.com/paper/2307.03784/) (2023)
- [MELAC: Massive Evaluation of Large Language Models with Alignment of Culture in Persian Language](https://mlsanity.com/paper/2508.00673/) (2025)

## Abstract (arXiv)

Hallucination is a persistent issue affecting all large language Models (LLMs), particularly within low-resource languages such as Persian. PerHalluEval (Persian Hallucination Evaluation) is the first dynamic hallucination evaluation benchmark tailored for the Persian language. Our benchmark leverages a three-stage LLM-driven pipeline, augmented with human validation, to generate plausible answers and summaries regarding QA and summarization tasks, focusing on detecting extrinsic and intrinsic hallucinations. Moreover, we used the log probabilities of generated tokens to select the most believable hallucinated instances. In addition, we engaged human annotators to highlight Persian-specific contexts in the QA dataset in order to evaluate LLMs' performance on content specifically related to Persian culture. Our evaluation of 12 LLMs, including open- and closed-source models using PerHalluEval, revealed that the models generally struggle in detecting hallucinated Persian text. We showed that providing external knowledge, i.e., the original document for the summarization task, could mitigate hallucination partially. Furthermore, there was no significant difference in terms of hallucination when comparing LLMs specifically trained for Persian with others.