Shaer: Controlled Arabic Poetry Generation with Meter Subform and Semantic Conditioning

arXiv cs.CL Papers
arabic-nlp poetry-generation controllable-generation qlora fine-tuning llm

Summary

Shaer: a controllable Classical Arabic poetry generation framework that jointly conditions on natural-language descriptions, meter subforms, and target hemistich counts, built by QLoRA fine-tuning Yehia-7B on a 116,032-poem enriched corpus, achieving 95.17% base-meter accuracy and strong length/subform control.

arXiv:2610.09756v1 Announce Type: new Abstract: Classical Arabic poetry generation requires simultaneously satisfying semantic, linguistic, and fine-grained prosodic constraints. Existing systems typically control broad poetic attributes but do not jointly model semantic intent, meter subform, and poem length. We present Shaer, a controllable Classical Arabic poetry generation framework jointly conditioned on natural-language descriptions, meter subforms, and target hemistich counts. To support this task, we construct an enriched corpus of 116,032 classical Arabic poems derived from Ashaar, containing normalized meter-subform labels and automatically generated, validated semantic descriptions. We then adapt Yehia-7B using QLoRA-based supervised fine-tuning with a completion-only objective. Our evaluation combines automatic assessment of base-meter conformity, requested-subform adherence, and length control with three LLM judges, blinded human evaluation, and memorization analysis. Shaer achieves 95.17% base-meter accuracy, 91.75% poem-level meter-subform accuracy, and 83.40% exact count accuracy. Relative to its untuned foundation model, these results represent gains of 68.68, 57.77, and 38.93 percentage points, respectively; Shaer also attains the highest base-meter accuracy among all evaluated systems. Multi-LLM evaluation and a blinded human assessment of top-ranked outputs further indicate competitive semantic and literary quality. Finally, analysis of all 3,481 test generations finds no exact copies from the training corpus or paired source poems. Code, models, and datasets are publicly available.
Original Article
View Cached Full Text

Cached at: 10/08/26, 12:45 PM 

# Shaer: Controlled Arabic Poetry Generation with Meter Subform and Semantic Conditioning
Source: [https://arxiv.org/abs/2610.09756](https://arxiv.org/abs/2610.09756)
[View PDF](https://arxiv.org/pdf/2610.09756)

> Abstract:Classical Arabic poetry generation requires simultaneously satisfying semantic, linguistic, and fine\-grained prosodic constraints\. Existing systems typically control broad poetic attributes but do not jointly model semantic intent, meter subform, and poem length\. We present Shaer, a controllable Classical Arabic poetry generation framework jointly conditioned on natural\-language descriptions, meter subforms, and target hemistich counts\. To support this task, we construct an enriched corpus of 116,032 classical Arabic poems derived from Ashaar, containing normalized meter\-subform labels and automatically generated, validated semantic descriptions\. We then adapt Yehia\-7B using QLoRA\-based supervised fine\-tuning with a completion\-only objective\. Our evaluation combines automatic assessment of base\-meter conformity, requested\-subform adherence, and length control with three LLM judges, blinded human evaluation, and memorization analysis\. Shaer achieves 95\.17% base\-meter accuracy, 91\.75% poem\-level meter\-subform accuracy, and 83\.40% exact count accuracy\. Relative to its untuned foundation model, these results represent gains of 68\.68, 57\.77, and 38\.93 percentage points, respectively; Shaer also attains the highest base\-meter accuracy among all evaluated systems\. Multi\-LLM evaluation and a blinded human assessment of top\-ranked outputs further indicate competitive semantic and literary quality\. Finally, analysis of all 3,481 test generations finds no exact copies from the training corpus or paired source poems\. Code, models, and datasets are publicly available\.

## Submission history

From: Ahmad Abbas \[[view email](https://arxiv.org/show-email/79a21def/2610.09756)\] **\[v1\]**Wed, 7 Oct 2026 09:44:21 UTC \(4,105 KB\)

Similar Articles