# Instruction-Guided Poetry Generation in Arabic and Its Dialects

arXiv: https://arxiv.org/abs/2604.27766 · HTML: https://mlsanity.com/paper/2604.27766/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 37 references that are in the map.
- First posted on arXiv 30 Apr 2026.

## Builds on (most cited first, 12 of 12)

- [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://mlsanity.com/paper/2105.13626/) (2021)
- [JudgeLM: Fine-tuned Large Language Models are Scalable Judges](https://mlsanity.com/paper/2310.17631/) (2023)
- [Help me write a poem: Instruction Tuning as a Vehicle for Collaborative Poetry Writing](https://mlsanity.com/paper/2210.13669/) (2022)
- [Fanar: An Arabic-Centric Multimodal Generative AI Platform](https://mlsanity.com/paper/2501.13944/) (2025)
- [ByGPT5: End-to-End Style-conditioned Poetry Generation with Token-free Language Models](https://mlsanity.com/paper/2212.10474/) (2022)
- [Chinese Poetry Generation with Planning based Neural Network](https://mlsanity.com/paper/1610.09889/) (2016)
- [PoeLM: A Meter- and Rhyme-Controllable Language Model for Unsupervised Poetry Generation](https://mlsanity.com/paper/2205.12206/) (2022)
- [CharPoet: A Chinese Classical Poetry Generation System Based on Token-free LLM](https://mlsanity.com/paper/2401.03512/) (2024)
- [Fann or Flop: A Multigenre, Multiera Benchmark for Arabic Poetry Understanding in LLMs](https://mlsanity.com/paper/2505.18152/) (2025)
- [LLM-based multi-agent poetry generation in non-cooperative environments](https://mlsanity.com/paper/2409.03659/) (2024)
- [AraPoemBERT: A Pretrained Language Model for Arabic Poetry Analysis](https://mlsanity.com/paper/2403.12392/) (2024)
- [A Rhythm-Aware Phrase Insertion for Classical Arabic Poetry Composition](https://mlsanity.com/paper/2509.18514/) (2025)

## Built on by (most cited first, 1 of 1)

- [Beyond Poetry: Can Large Language Models Generate Classical Arabic Maqamat?](https://mlsanity.com/paper/2609.28245/) (2026)

## Abstract (arXiv)

Poetry has long been a central art form for Arabic speakers, serving as a powerful medium of expression and cultural identity. While modern Arabic speakers continue to value poetry, existing research on Arabic poetry within Large Language Models (LLMs) has primarily focused on analysis tasks such as interpretation or metadata prediction, e.g., rhyme schemes and titles. In contrast, our work addresses the practical aspect of poetry creation in Arabic by introducing controllable generation capabilities to assist users in writing poetry. Specifically, we present a large-scale, carefully curated instruction-based dataset in Modern Standard Arabic (MSA) and various Arabic dialects. This dataset enables tasks such as writing, revising, and continuing poems based on predefined criteria, including style and rhyme, as well as performing poetry analysis. Our experiments show that fine-tuning LLMs on this dataset yields models that can effectively generate poetry that is aligned with user requirements, based on both automated metrics and human evaluation with native Arabic speakers. The data and the code are available at https://github.com/mbzuai-nlp/instructpoet-ar