@inproceedings{srivastava-etal-2024-evaluating,
title = "Evaluating {LLM}s' Mathematical Reasoning in Financial Document Question Answering",
author = "Srivastava, Pragya and
Malik, Manuj and
Gupta, Vivek and
Ganu, Tanuja and
Roth, Dan",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.231/",
doi = "10.18653/v1/2024.findings-acl.231",
pages = "3853--3878",
abstract = "Large Language Models (LLMs), excel in natural language understanding, but their capability for complex mathematical reasoning with a hybrid of structured tables and unstructured text remain uncertain. This study explores LLMs' mathematical reasoning on four financial tabular question-answering datasets: TATQA, FinQA, ConvFinQA, and Multihiertt. Through extensive experiments with various models and prompting techniques, we assess how LLMs adapt to complex tables and mathematical tasks. We focus on sensitivity to table complexity and performance variations with an increasing number of arithmetic reasoning steps. The results provide insights into LLMs' capabilities and limitations in handling complex mathematical scenarios for semi-structured tables. Ultimately, we introduce a novel prompting technique EEDP tailored to semi-structured documents, matching or outperforming baselines performance while providing a nuanced understanding of LLMs abilities."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="srivastava-etal-2024-evaluating">
<titleInfo>
<title>Evaluating LLMs’ Mathematical Reasoning in Financial Document Question Answering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pragya</namePart>
<namePart type="family">Srivastava</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manuj</namePart>
<namePart type="family">Malik</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanuja</namePart>
<namePart type="family">Ganu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dan</namePart>
<namePart type="family">Roth</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large Language Models (LLMs), excel in natural language understanding, but their capability for complex mathematical reasoning with a hybrid of structured tables and unstructured text remain uncertain. This study explores LLMs’ mathematical reasoning on four financial tabular question-answering datasets: TATQA, FinQA, ConvFinQA, and Multihiertt. Through extensive experiments with various models and prompting techniques, we assess how LLMs adapt to complex tables and mathematical tasks. We focus on sensitivity to table complexity and performance variations with an increasing number of arithmetic reasoning steps. The results provide insights into LLMs’ capabilities and limitations in handling complex mathematical scenarios for semi-structured tables. Ultimately, we introduce a novel prompting technique EEDP tailored to semi-structured documents, matching or outperforming baselines performance while providing a nuanced understanding of LLMs abilities.</abstract>
<identifier type="citekey">srivastava-etal-2024-evaluating</identifier>
<identifier type="doi">10.18653/v1/2024.findings-acl.231</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.231/</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>3853</start>
<end>3878</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Evaluating LLMs’ Mathematical Reasoning in Financial Document Question Answering
%A Srivastava, Pragya
%A Malik, Manuj
%A Gupta, Vivek
%A Ganu, Tanuja
%A Roth, Dan
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics: ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F srivastava-etal-2024-evaluating
%X Large Language Models (LLMs), excel in natural language understanding, but their capability for complex mathematical reasoning with a hybrid of structured tables and unstructured text remain uncertain. This study explores LLMs’ mathematical reasoning on four financial tabular question-answering datasets: TATQA, FinQA, ConvFinQA, and Multihiertt. Through extensive experiments with various models and prompting techniques, we assess how LLMs adapt to complex tables and mathematical tasks. We focus on sensitivity to table complexity and performance variations with an increasing number of arithmetic reasoning steps. The results provide insights into LLMs’ capabilities and limitations in handling complex mathematical scenarios for semi-structured tables. Ultimately, we introduce a novel prompting technique EEDP tailored to semi-structured documents, matching or outperforming baselines performance while providing a nuanced understanding of LLMs abilities.
%R 10.18653/v1/2024.findings-acl.231
%U https://aclanthology.org/2024.findings-acl.231/
%U https://doi.org/10.18653/v1/2024.findings-acl.231
%P 3853-3878
Markdown (Informal)
[Evaluating LLMs’ Mathematical Reasoning in Financial Document Question Answering](https://aclanthology.org/2024.findings-acl.231/) (Srivastava et al., Findings 2024)
ACL