@inproceedings{suppa-etal-2026-ragthoven,
title = "{RAG}thoven at {S}em{E}val-2026 Task 1: A Multi-Stage Pipeline Walks Into a Benchmark and Barely Clears the Bar",
author = "Suppa, Marek and
Ondrejov{\'a}, Vikt{\'o}ria and
Ganajov{\'a}, Lucia and
Karetka, Gregor and
Skala, Daniel",
editor = "Kochmar, Ekaterina and
Ghosh, Debanjan and
North, Kai and
Komachi, Mamoru",
booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.semeval-1.416/",
pages = "3343--3356",
ISBN = "979-8-89176-414-9",
abstract = "We present {\textbackslash}textsc{\{}RAGthoven{\}}, our system for SemEval-2026 Task{\textasciitilde}1 (MuWaHaHa), Subtask{\textasciitilde}A (multilingual constrained humor generation in English, Spanish, and Chinese).{\textbackslash}textsc{\{}RAGthoven{\}} decomposes creative text generation into a multi-stage large language model (LLM) pipeline ({\textbackslash}textit{\{}Planner{\}}, {\textbackslash}textit{\{}Writer{\}}, {\textbackslash}textit{\{}Reflector{\}}, {\textbackslash}textit{\{}Judge{\}}) grounded in computational humor theories (Benign Violation Theory, Script-based Semantic Theory of Humor) and iteratively refined through prompt engineering across ten experiments.In our final configuration, we augment the Planner with retrieval-augmented generation (RAG) from a curated joke corpus, seeding generation with diverse joke mechanisms.We additionally explore an agentic variant that exposes the same four pipeline stages as tool-calling agents orchestrated by a model loop with a {\textbackslash}textsc{\{}ConstraintAudit{\}} checker. While it achieves full constraint compliance, human pairwise evaluation did not reveal a significant quality advantage over the simpler non-agentic baseline.{\textbackslash}textsc{\{}RAGthoven{\}} achieves Rank{\textasciitilde}1 in all three languages, with the strongest result in Spanish (Elo 1182, 42 points above the Gemini{\textasciitilde}2.5{\textasciitilde}Flash baseline).However, while the system leads in raw Elo in Spanish, it shares Rank{\textasciitilde}1 with the baseline in all three languages due to overlapping confidence intervals; in English and Chinese the gap narrows further, suggesting that elaborate multi-stage prompt engineering may offer diminishing returns once a strong frontier model is in the loop."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="suppa-etal-2026-ragthoven">
<titleInfo>
<title>RAGthoven at SemEval-2026 Task 1: A Multi-Stage Pipeline Walks Into a Benchmark and Barely Clears the Bar</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marek</namePart>
<namePart type="family">Suppa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viktória</namePart>
<namePart type="family">Ondrejová</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Ganajová</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gregor</namePart>
<namePart type="family">Karetka</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Skala</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 20th International Workshop on Semantic Evaluation (2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debanjan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">North</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-414-9</identifier>
</relatedItem>
<abstract>We present \textbackslashtextsc{RAGthoven}, our system for SemEval-2026 Task~1 (MuWaHaHa), Subtask~A (multilingual constrained humor generation in English, Spanish, and Chinese).\textbackslashtextsc{RAGthoven} decomposes creative text generation into a multi-stage large language model (LLM) pipeline (\textbackslashtextit{Planner}, \textbackslashtextit{Writer}, \textbackslashtextit{Reflector}, \textbackslashtextit{Judge}) grounded in computational humor theories (Benign Violation Theory, Script-based Semantic Theory of Humor) and iteratively refined through prompt engineering across ten experiments.In our final configuration, we augment the Planner with retrieval-augmented generation (RAG) from a curated joke corpus, seeding generation with diverse joke mechanisms.We additionally explore an agentic variant that exposes the same four pipeline stages as tool-calling agents orchestrated by a model loop with a \textbackslashtextsc{ConstraintAudit} checker. While it achieves full constraint compliance, human pairwise evaluation did not reveal a significant quality advantage over the simpler non-agentic baseline.\textbackslashtextsc{RAGthoven} achieves Rank~1 in all three languages, with the strongest result in Spanish (Elo 1182, 42 points above the Gemini~2.5~Flash baseline).However, while the system leads in raw Elo in Spanish, it shares Rank~1 with the baseline in all three languages due to overlapping confidence intervals; in English and Chinese the gap narrows further, suggesting that elaborate multi-stage prompt engineering may offer diminishing returns once a strong frontier model is in the loop.</abstract>
<identifier type="citekey">suppa-etal-2026-ragthoven</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.416/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>3343</start>
<end>3356</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T RAGthoven at SemEval-2026 Task 1: A Multi-Stage Pipeline Walks Into a Benchmark and Barely Clears the Bar
%A Suppa, Marek
%A Ondrejová, Viktória
%A Ganajová, Lucia
%A Karetka, Gregor
%A Skala, Daniel
%Y Kochmar, Ekaterina
%Y Ghosh, Debanjan
%Y North, Kai
%Y Komachi, Mamoru
%S Proceedings of the 20th International Workshop on Semantic Evaluation (2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-414-9
%F suppa-etal-2026-ragthoven
%X We present \textbackslashtextsc{RAGthoven}, our system for SemEval-2026 Task~1 (MuWaHaHa), Subtask~A (multilingual constrained humor generation in English, Spanish, and Chinese).\textbackslashtextsc{RAGthoven} decomposes creative text generation into a multi-stage large language model (LLM) pipeline (\textbackslashtextit{Planner}, \textbackslashtextit{Writer}, \textbackslashtextit{Reflector}, \textbackslashtextit{Judge}) grounded in computational humor theories (Benign Violation Theory, Script-based Semantic Theory of Humor) and iteratively refined through prompt engineering across ten experiments.In our final configuration, we augment the Planner with retrieval-augmented generation (RAG) from a curated joke corpus, seeding generation with diverse joke mechanisms.We additionally explore an agentic variant that exposes the same four pipeline stages as tool-calling agents orchestrated by a model loop with a \textbackslashtextsc{ConstraintAudit} checker. While it achieves full constraint compliance, human pairwise evaluation did not reveal a significant quality advantage over the simpler non-agentic baseline.\textbackslashtextsc{RAGthoven} achieves Rank~1 in all three languages, with the strongest result in Spanish (Elo 1182, 42 points above the Gemini~2.5~Flash baseline).However, while the system leads in raw Elo in Spanish, it shares Rank~1 with the baseline in all three languages due to overlapping confidence intervals; in English and Chinese the gap narrows further, suggesting that elaborate multi-stage prompt engineering may offer diminishing returns once a strong frontier model is in the loop.
%U https://aclanthology.org/2026.semeval-1.416/
%P 3343-3356Markdown (Informal)
[RAGthoven at SemEval-2026 Task 1: A Multi-Stage Pipeline Walks Into a Benchmark and Barely Clears the Bar](https://aclanthology.org/2026.semeval-1.416/) (Suppa et al., SemEval 2026)
ACL