@inproceedings{shahrouri-etal-2026-leedsmeng26,
title = "{L}eeds{ME}ng26: Qwen + Gemini for {F}in{C}ausal 2026 Causality Detection in Financial Narrative Texts",
author = "Shahrouri, Zaid and
Ivienagbor, Ayomide and
Asad, Idrees and
Shrestha, Rijul and
Bal, Yasemin and
Nadeem, Zahaab",
editor = "El-Haj, Mo and
Moreno Sandoval, Antonio and
Garcia-Serrano, Ana and
Chen, Chung-Chi and
Rayson, Paul and
Torterolo Orta, Yanco Amor and
Martinez, Paloma and
Porta, Jordi",
booktitle = "The 7th Financial Narrative Processing Workshop",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "European Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.fnp-1.17/",
doi = "10.63317/4aaj7cwvpjxy",
pages = "160--168",
abstract = "This paper presents the LeedsMEng26 system for the FinCausal 2026 shared task (CITATION) on financial causality detection in narrative texts. The task is formulated as extractive question answering over English and Spanish financial reports, where systems must return a verbatim span from the context that answers an abstractive question about a cause or an effect. We propose a two-stage pipeline consisting of candidate span generation followed by span verification and boundary refinement under a strict extractiveness constraint. We evaluate both an extractive RoBERTa-based baseline and instruction-tuned large language models. Results show that Qwen-2.5-1.5B-Instruct is a stronger candidate generator than the RoBERTa baseline, and that a second-stage verifier further improves answer boundary accuracy and overall adequacy. Our best configuration, Qwen-2.5-1.5B-Instruct with Gemini-2.5-flash refinement, achieved an adequacy score of 4.7000 for English and 4.6143 for Spanish. These findings suggest that a modular generation-and-verification pipeline is effective for extractive financial causality detection."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shahrouri-etal-2026-leedsmeng26">
<titleInfo>
<title>LeedsMEng26: Qwen + Gemini for FinCausal 2026 Causality Detection in Financial Narrative Texts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zaid</namePart>
<namePart type="family">Shahrouri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayomide</namePart>
<namePart type="family">Ivienagbor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Idrees</namePart>
<namePart type="family">Asad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rijul</namePart>
<namePart type="family">Shrestha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yasemin</namePart>
<namePart type="family">Bal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zahaab</namePart>
<namePart type="family">Nadeem</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>The 7th Financial Narrative Processing Workshop</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mo</namePart>
<namePart type="family">El-Haj</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Moreno Sandoval</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ana</namePart>
<namePart type="family">Garcia-Serrano</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chung-Chi</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">Rayson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanco</namePart>
<namePart type="given">Amor</namePart>
<namePart type="family">Torterolo Orta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paloma</namePart>
<namePart type="family">Martinez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordi</namePart>
<namePart type="family">Porta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>European Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper presents the LeedsMEng26 system for the FinCausal 2026 shared task (CITATION) on financial causality detection in narrative texts. The task is formulated as extractive question answering over English and Spanish financial reports, where systems must return a verbatim span from the context that answers an abstractive question about a cause or an effect. We propose a two-stage pipeline consisting of candidate span generation followed by span verification and boundary refinement under a strict extractiveness constraint. We evaluate both an extractive RoBERTa-based baseline and instruction-tuned large language models. Results show that Qwen-2.5-1.5B-Instruct is a stronger candidate generator than the RoBERTa baseline, and that a second-stage verifier further improves answer boundary accuracy and overall adequacy. Our best configuration, Qwen-2.5-1.5B-Instruct with Gemini-2.5-flash refinement, achieved an adequacy score of 4.7000 for English and 4.6143 for Spanish. These findings suggest that a modular generation-and-verification pipeline is effective for extractive financial causality detection.</abstract>
<identifier type="citekey">shahrouri-etal-2026-leedsmeng26</identifier>
<identifier type="doi">10.63317/4aaj7cwvpjxy</identifier>
<location>
<url>https://aclanthology.org/2026.fnp-1.17/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>160</start>
<end>168</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LeedsMEng26: Qwen + Gemini for FinCausal 2026 Causality Detection in Financial Narrative Texts
%A Shahrouri, Zaid
%A Ivienagbor, Ayomide
%A Asad, Idrees
%A Shrestha, Rijul
%A Bal, Yasemin
%A Nadeem, Zahaab
%Y El-Haj, Mo
%Y Moreno Sandoval, Antonio
%Y Garcia-Serrano, Ana
%Y Chen, Chung-Chi
%Y Rayson, Paul
%Y Torterolo Orta, Yanco Amor
%Y Martinez, Paloma
%Y Porta, Jordi
%S The 7th Financial Narrative Processing Workshop
%D 2026
%8 May
%I European Language Resources Association (ELRA)
%C Palma de Mallorca, Spain
%F shahrouri-etal-2026-leedsmeng26
%X This paper presents the LeedsMEng26 system for the FinCausal 2026 shared task (CITATION) on financial causality detection in narrative texts. The task is formulated as extractive question answering over English and Spanish financial reports, where systems must return a verbatim span from the context that answers an abstractive question about a cause or an effect. We propose a two-stage pipeline consisting of candidate span generation followed by span verification and boundary refinement under a strict extractiveness constraint. We evaluate both an extractive RoBERTa-based baseline and instruction-tuned large language models. Results show that Qwen-2.5-1.5B-Instruct is a stronger candidate generator than the RoBERTa baseline, and that a second-stage verifier further improves answer boundary accuracy and overall adequacy. Our best configuration, Qwen-2.5-1.5B-Instruct with Gemini-2.5-flash refinement, achieved an adequacy score of 4.7000 for English and 4.6143 for Spanish. These findings suggest that a modular generation-and-verification pipeline is effective for extractive financial causality detection.
%R 10.63317/4aaj7cwvpjxy
%U https://aclanthology.org/2026.fnp-1.17/
%U https://doi.org/10.63317/4aaj7cwvpjxy
%P 160-168
Markdown (Informal)
[LeedsMEng26: Qwen + Gemini for FinCausal 2026 Causality Detection in Financial Narrative Texts](https://aclanthology.org/2026.fnp-1.17/) (Shahrouri et al., FNP 2026)
ACL