@inproceedings{lopez-gude-etal-2025-lys,
title = "{L}y{S} at {S}em{E}val 2025 Task 8: Zero-Shot Code Generation for Tabular {QA}",
author = "L{\'o}pez Gude, Adri{\'a}n and
Santos R{\'i}os, Roi and
Prado Vali{\~n}o, Francisco and
Ezquerro, Ana and
Vilares, Jes{\'u}s",
editor = "Rosenthal, Sara and
Ros{\'a}, Aiala and
Ghosh, Debanjan and
Zampieri, Marcos",
booktitle = "Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.semeval-1.171/",
pages = "1282--1288",
ISBN = "979-8-89176-273-2",
abstract = "We developed a zero-shot pipeline that leverages an Large Language Model to generate functional code capable of extracting the relevant information from tabular data based on an input question. Our approach consists of a modular pipeline where the main code generator module is supported by additional components that identify the most relevant columns and analyze their data types to improve extraction accuracy. In the event that the generated code fails, an iterative refinement process is triggered, incorporating the error feedback into a new generation prompt to enhance robustness. Our results show that zero-shot code generation is a valid approach for Tabular QA, achieving rank 33 of 53 in the test phase despite the lack of task-specific fine-tuning."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lopez-gude-etal-2025-lys">
<titleInfo>
<title>LyS at SemEval 2025 Task 8: Zero-Shot Code Generation for Tabular QA</title>
</titleInfo>
<name type="personal">
<namePart type="given">Adrián</namePart>
<namePart type="family">López Gude</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roi</namePart>
<namePart type="family">Santos Ríos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Francisco</namePart>
<namePart type="family">Prado Valiño</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ana</namePart>
<namePart type="family">Ezquerro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jesús</namePart>
<namePart type="family">Vilares</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sara</namePart>
<namePart type="family">Rosenthal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aiala</namePart>
<namePart type="family">Rosá</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">Marcos</namePart>
<namePart type="family">Zampieri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-273-2</identifier>
</relatedItem>
<abstract>We developed a zero-shot pipeline that leverages an Large Language Model to generate functional code capable of extracting the relevant information from tabular data based on an input question. Our approach consists of a modular pipeline where the main code generator module is supported by additional components that identify the most relevant columns and analyze their data types to improve extraction accuracy. In the event that the generated code fails, an iterative refinement process is triggered, incorporating the error feedback into a new generation prompt to enhance robustness. Our results show that zero-shot code generation is a valid approach for Tabular QA, achieving rank 33 of 53 in the test phase despite the lack of task-specific fine-tuning.</abstract>
<identifier type="citekey">lopez-gude-etal-2025-lys</identifier>
<location>
<url>https://aclanthology.org/2025.semeval-1.171/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>1282</start>
<end>1288</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LyS at SemEval 2025 Task 8: Zero-Shot Code Generation for Tabular QA
%A López Gude, Adrián
%A Santos Ríos, Roi
%A Prado Valiño, Francisco
%A Ezquerro, Ana
%A Vilares, Jesús
%Y Rosenthal, Sara
%Y Rosá, Aiala
%Y Ghosh, Debanjan
%Y Zampieri, Marcos
%S Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-273-2
%F lopez-gude-etal-2025-lys
%X We developed a zero-shot pipeline that leverages an Large Language Model to generate functional code capable of extracting the relevant information from tabular data based on an input question. Our approach consists of a modular pipeline where the main code generator module is supported by additional components that identify the most relevant columns and analyze their data types to improve extraction accuracy. In the event that the generated code fails, an iterative refinement process is triggered, incorporating the error feedback into a new generation prompt to enhance robustness. Our results show that zero-shot code generation is a valid approach for Tabular QA, achieving rank 33 of 53 in the test phase despite the lack of task-specific fine-tuning.
%U https://aclanthology.org/2025.semeval-1.171/
%P 1282-1288
Markdown (Informal)
[LyS at SemEval 2025 Task 8: Zero-Shot Code Generation for Tabular QA](https://aclanthology.org/2025.semeval-1.171/) (López Gude et al., SemEval 2025)
ACL