@inproceedings{chatzikyriakidis-natsina-2026-llms-got,
title = "{LLM}s Got Rhyme? Hybrid Phonological Filtering for {G}reek Poetry Rhyme Detection and Generation",
author = "Chatzikyriakidis, Stergios and
Natsina, Anastasia",
editor = "Alves, Diego and
Bizzoni, Yuri and
Degaetano-Ortlieb, Stefania and
Kazantseva, Anna and
Pagel, Janis and
Szpakowicz, Stan",
booktitle = "Proceedings of the 10th Joint {SIGHUM} Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature 2026",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.latechclfl-1.9/",
pages = "87--101",
ISBN = "979-8-89176-373-9",
abstract = "Large Language Models (LLMs), even though exhibiting multiple capabilities on many NLP tasks, struggle with phonologically-grounded phenomena like rhyme detection and generation. When one moves to lower-resource languages such as Modern Greek, this is even more evident. In this paper, we present a hybrid neural-symbolic system that combines LLMs with deterministic phonological algorithms to achieve accurate rhyme identification and generation. We implement a comprehensive taxonomy of Greek rhyme types and employ an agentic generation pipeline with phonological verification. We use multiple prompting strategies (zero-shot, few-shot, Chain-of-Thought, and RAG-augmented) across several LLMs including Claude 3.7 and 4.5, GPT-4o, Gemini 2.0 and open-weight models like Llama 3.1 8B and 70B and Mistral Large. Results reveal a significant reasoning gap: while native-like models (Claude 3.7) perform intuitively (40{\textbackslash}{\%} accuracy in identification), reasoning-heavy models (Claude 4.5) achieve state-of-the-art performance (54{\textbackslash}{\%}) only when prompted with Chain-of-Thought. Most critically, pure LLM generation fails significantly (under 4{\textbackslash}{\%} valid poems), while our hybrid verification loop restores performance to 73.1{\textbackslash}{\%}. Along with the system presented, we further release a corpus of 40,000+ rhymes, derived from the {\textbackslash}textit{\{}Anemoskala{\}} and {\textbackslash}textit{\{}Interwar Poetry{\}} corpora, to support future research."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chatzikyriakidis-natsina-2026-llms-got">
<titleInfo>
<title>LLMs Got Rhyme? Hybrid Phonological Filtering for Greek Poetry Rhyme Detection and Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stergios</namePart>
<namePart type="family">Chatzikyriakidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anastasia</namePart>
<namePart type="family">Natsina</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 10th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Diego</namePart>
<namePart type="family">Alves</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuri</namePart>
<namePart type="family">Bizzoni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stefania</namePart>
<namePart type="family">Degaetano-Ortlieb</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Kazantseva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Janis</namePart>
<namePart type="family">Pagel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stan</namePart>
<namePart type="family">Szpakowicz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-373-9</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs), even though exhibiting multiple capabilities on many NLP tasks, struggle with phonologically-grounded phenomena like rhyme detection and generation. When one moves to lower-resource languages such as Modern Greek, this is even more evident. In this paper, we present a hybrid neural-symbolic system that combines LLMs with deterministic phonological algorithms to achieve accurate rhyme identification and generation. We implement a comprehensive taxonomy of Greek rhyme types and employ an agentic generation pipeline with phonological verification. We use multiple prompting strategies (zero-shot, few-shot, Chain-of-Thought, and RAG-augmented) across several LLMs including Claude 3.7 and 4.5, GPT-4o, Gemini 2.0 and open-weight models like Llama 3.1 8B and 70B and Mistral Large. Results reveal a significant reasoning gap: while native-like models (Claude 3.7) perform intuitively (40\textbackslash% accuracy in identification), reasoning-heavy models (Claude 4.5) achieve state-of-the-art performance (54\textbackslash%) only when prompted with Chain-of-Thought. Most critically, pure LLM generation fails significantly (under 4\textbackslash% valid poems), while our hybrid verification loop restores performance to 73.1\textbackslash%. Along with the system presented, we further release a corpus of 40,000+ rhymes, derived from the \textbackslashtextit{Anemoskala} and \textbackslashtextit{Interwar Poetry} corpora, to support future research.</abstract>
<identifier type="citekey">chatzikyriakidis-natsina-2026-llms-got</identifier>
<location>
<url>https://aclanthology.org/2026.latechclfl-1.9/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>87</start>
<end>101</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LLMs Got Rhyme? Hybrid Phonological Filtering for Greek Poetry Rhyme Detection and Generation
%A Chatzikyriakidis, Stergios
%A Natsina, Anastasia
%Y Alves, Diego
%Y Bizzoni, Yuri
%Y Degaetano-Ortlieb, Stefania
%Y Kazantseva, Anna
%Y Pagel, Janis
%Y Szpakowicz, Stan
%S Proceedings of the 10th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature 2026
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-373-9
%F chatzikyriakidis-natsina-2026-llms-got
%X Large Language Models (LLMs), even though exhibiting multiple capabilities on many NLP tasks, struggle with phonologically-grounded phenomena like rhyme detection and generation. When one moves to lower-resource languages such as Modern Greek, this is even more evident. In this paper, we present a hybrid neural-symbolic system that combines LLMs with deterministic phonological algorithms to achieve accurate rhyme identification and generation. We implement a comprehensive taxonomy of Greek rhyme types and employ an agentic generation pipeline with phonological verification. We use multiple prompting strategies (zero-shot, few-shot, Chain-of-Thought, and RAG-augmented) across several LLMs including Claude 3.7 and 4.5, GPT-4o, Gemini 2.0 and open-weight models like Llama 3.1 8B and 70B and Mistral Large. Results reveal a significant reasoning gap: while native-like models (Claude 3.7) perform intuitively (40\textbackslash% accuracy in identification), reasoning-heavy models (Claude 4.5) achieve state-of-the-art performance (54\textbackslash%) only when prompted with Chain-of-Thought. Most critically, pure LLM generation fails significantly (under 4\textbackslash% valid poems), while our hybrid verification loop restores performance to 73.1\textbackslash%. Along with the system presented, we further release a corpus of 40,000+ rhymes, derived from the \textbackslashtextit{Anemoskala} and \textbackslashtextit{Interwar Poetry} corpora, to support future research.
%U https://aclanthology.org/2026.latechclfl-1.9/
%P 87-101Markdown (Informal)
[LLMs Got Rhyme? Hybrid Phonological Filtering for Greek Poetry Rhyme Detection and Generation](https://aclanthology.org/2026.latechclfl-1.9/) (Chatzikyriakidis & Natsina, LaTeCH-CLfL 2026)
ACL