@inproceedings{yone-etal-2026-reasoning,
title = "Reasoning Graph-Structured Question Answering: Datasets and Insights from {LLM} Benchmarking",
author = "Yone, Khin and
Trivedi, Devasha and
Pahilajani, Anish and
Shuai, Jincen and
Jain, Samyak Rajesh and
Rossi, Ryan and
Ahmed, Nesreen K. and
Dernoncourt, Franck and
Wang, Yu and
Park, Namyong",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.414/",
doi = "10.63317/4zjpmtqxxtx4",
pages = "5301--5316",
abstract = "Large Language Models (LLMs) have shown remarkable success in multi-hop question-answering (M-QA) due to their advanced reasoning capabilities. However, the influence of reasoning structures on their performance remains underexplored, primarily due to the lack of M-QA datasets that explicitly encode the reasoning pathways underlying each question-answer pair. To address this gap, we introduce the reasoning graph-structured question answering dataset (GRS-QA), which provides both semantic contexts and reasoning structures for the QA pairs. Unlike existing M-QA datasets, GRS-QA explicitly captures intricate reasoning pathways through reasoning graphs, where nodes correspond to textual contexts and edges denote logical flows. Using GRS-QA, we systematically evaluate LLM performance across varying context structures, prompting styles, and data domains. Our empirical analysis reveals that LLMs perform differently based on the reasoning structure, context, and prompting styles, indicating their varying ability to leverage graph-structured knowledge. Notably, providing explicit reasoning guidance proves more effective than supplying contextual information alone."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yone-etal-2026-reasoning">
<titleInfo>
<title>Reasoning Graph-Structured Question Answering: Datasets and Insights from LLM Benchmarking</title>
</titleInfo>
<name type="personal">
<namePart type="given">Khin</namePart>
<namePart type="family">Yone</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Devasha</namePart>
<namePart type="family">Trivedi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anish</namePart>
<namePart type="family">Pahilajani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jincen</namePart>
<namePart type="family">Shuai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Samyak</namePart>
<namePart type="given">Rajesh</namePart>
<namePart type="family">Jain</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryan</namePart>
<namePart type="family">Rossi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nesreen</namePart>
<namePart type="given">K</namePart>
<namePart type="family">Ahmed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Franck</namePart>
<namePart type="family">Dernoncourt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Namyong</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large Language Models (LLMs) have shown remarkable success in multi-hop question-answering (M-QA) due to their advanced reasoning capabilities. However, the influence of reasoning structures on their performance remains underexplored, primarily due to the lack of M-QA datasets that explicitly encode the reasoning pathways underlying each question-answer pair. To address this gap, we introduce the reasoning graph-structured question answering dataset (GRS-QA), which provides both semantic contexts and reasoning structures for the QA pairs. Unlike existing M-QA datasets, GRS-QA explicitly captures intricate reasoning pathways through reasoning graphs, where nodes correspond to textual contexts and edges denote logical flows. Using GRS-QA, we systematically evaluate LLM performance across varying context structures, prompting styles, and data domains. Our empirical analysis reveals that LLMs perform differently based on the reasoning structure, context, and prompting styles, indicating their varying ability to leverage graph-structured knowledge. Notably, providing explicit reasoning guidance proves more effective than supplying contextual information alone.</abstract>
<identifier type="citekey">yone-etal-2026-reasoning</identifier>
<identifier type="doi">10.63317/4zjpmtqxxtx4</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.414/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>5301</start>
<end>5316</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Reasoning Graph-Structured Question Answering: Datasets and Insights from LLM Benchmarking
%A Yone, Khin
%A Trivedi, Devasha
%A Pahilajani, Anish
%A Shuai, Jincen
%A Jain, Samyak Rajesh
%A Rossi, Ryan
%A Ahmed, Nesreen K.
%A Dernoncourt, Franck
%A Wang, Yu
%A Park, Namyong
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F yone-etal-2026-reasoning
%X Large Language Models (LLMs) have shown remarkable success in multi-hop question-answering (M-QA) due to their advanced reasoning capabilities. However, the influence of reasoning structures on their performance remains underexplored, primarily due to the lack of M-QA datasets that explicitly encode the reasoning pathways underlying each question-answer pair. To address this gap, we introduce the reasoning graph-structured question answering dataset (GRS-QA), which provides both semantic contexts and reasoning structures for the QA pairs. Unlike existing M-QA datasets, GRS-QA explicitly captures intricate reasoning pathways through reasoning graphs, where nodes correspond to textual contexts and edges denote logical flows. Using GRS-QA, we systematically evaluate LLM performance across varying context structures, prompting styles, and data domains. Our empirical analysis reveals that LLMs perform differently based on the reasoning structure, context, and prompting styles, indicating their varying ability to leverage graph-structured knowledge. Notably, providing explicit reasoning guidance proves more effective than supplying contextual information alone.
%R 10.63317/4zjpmtqxxtx4
%U https://aclanthology.org/2026.lrec-1.414/
%U https://doi.org/10.63317/4zjpmtqxxtx4
%P 5301-5316
Markdown (Informal)
[Reasoning Graph-Structured Question Answering: Datasets and Insights from LLM Benchmarking](https://aclanthology.org/2026.lrec-1.414/) (Yone et al., LREC 2026)
ACL
- Khin Yone, Devasha Trivedi, Anish Pahilajani, Jincen Shuai, Samyak Rajesh Jain, Ryan Rossi, Nesreen K. Ahmed, Franck Dernoncourt, Yu Wang, and Namyong Park. 2026. Reasoning Graph-Structured Question Answering: Datasets and Insights from LLM Benchmarking. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 5301–5316, Palma de Mallorca, Spain. ELRA Language Resource Association.