@inproceedings{vlachos-etal-2025-building,
title = "Building Open-Retrieval Conversational Question Answering Systems by Generating Synthetic Data and Decontextualizing User Questions",
author = "Vlachos, Christos and
Stylianou, Nikolaos and
Fiotaki, Alexandra and
Methenitis, Spiros and
Palogiannidi, Elisavet and
Stafylakis, Themos and
Androutsopoulos, Ion",
editor = "B{\'e}chet, Fr{\'e}d{\'e}ric and
Lef{\`e}vre, Fabrice and
Asher, Nicholas and
Kim, Seokhwan and
Merlin, Teva",
booktitle = "Proceedings of the 26th Annual Meeting of the Special Interest Group on Discourse and Dialogue",
month = aug,
year = "2025",
address = "Avignon, France",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.sigdial-1.52/",
pages = "641--657",
abstract = "We consider open-retrieval conversational question answering (OR-CONVQA), an extension of question answering where system responses need to be (i) aware of dialog history and (ii) grounded in documents (or document fragments) retrieved per question. Domain-specific OR-CONVQA training datasets are crucial for real-world applications, but hard to obtain. We propose a pipeline that capitalizes on the abundance of plain text documents in organizations (e.g., product documentation) to automatically produce realistic OR-CONVQA dialogs with annotations. Similarly to real-world humanannotated OR-CONVQA datasets, we generate in-dialog question-answer pairs, self-contained (decontextualized, e.g., no referring expressions) versions of user questions, and propositions (sentences expressing prominent information from the documents) the system responses are grounded in. We show how the synthetic dialogs can be used to train efficient question rewriters that decontextualize user questions, allowing existing dialog-unaware retrievers to be utilized. The retrieved information and the decontextualized question are then passed on to an LLM that generates the system{'}s response."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vlachos-etal-2025-building">
<titleInfo>
<title>Building Open-Retrieval Conversational Question Answering Systems by Generating Synthetic Data and Decontextualizing User Questions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolaos</namePart>
<namePart type="family">Stylianou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexandra</namePart>
<namePart type="family">Fiotaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Spiros</namePart>
<namePart type="family">Methenitis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Elisavet</namePart>
<namePart type="family">Palogiannidi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Themos</namePart>
<namePart type="family">Stafylakis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ion</namePart>
<namePart type="family">Androutsopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 26th Annual Meeting of the Special Interest Group on Discourse and Dialogue</title>
</titleInfo>
<name type="personal">
<namePart type="given">Frédéric</namePart>
<namePart type="family">Béchet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabrice</namePart>
<namePart type="family">Lefèvre</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicholas</namePart>
<namePart type="family">Asher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seokhwan</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Teva</namePart>
<namePart type="family">Merlin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Avignon, France</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We consider open-retrieval conversational question answering (OR-CONVQA), an extension of question answering where system responses need to be (i) aware of dialog history and (ii) grounded in documents (or document fragments) retrieved per question. Domain-specific OR-CONVQA training datasets are crucial for real-world applications, but hard to obtain. We propose a pipeline that capitalizes on the abundance of plain text documents in organizations (e.g., product documentation) to automatically produce realistic OR-CONVQA dialogs with annotations. Similarly to real-world humanannotated OR-CONVQA datasets, we generate in-dialog question-answer pairs, self-contained (decontextualized, e.g., no referring expressions) versions of user questions, and propositions (sentences expressing prominent information from the documents) the system responses are grounded in. We show how the synthetic dialogs can be used to train efficient question rewriters that decontextualize user questions, allowing existing dialog-unaware retrievers to be utilized. The retrieved information and the decontextualized question are then passed on to an LLM that generates the system’s response.</abstract>
<identifier type="citekey">vlachos-etal-2025-building</identifier>
<location>
<url>https://aclanthology.org/2025.sigdial-1.52/</url>
</location>
<part>
<date>2025-08</date>
<extent unit="page">
<start>641</start>
<end>657</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Building Open-Retrieval Conversational Question Answering Systems by Generating Synthetic Data and Decontextualizing User Questions
%A Vlachos, Christos
%A Stylianou, Nikolaos
%A Fiotaki, Alexandra
%A Methenitis, Spiros
%A Palogiannidi, Elisavet
%A Stafylakis, Themos
%A Androutsopoulos, Ion
%Y Béchet, Frédéric
%Y Lefèvre, Fabrice
%Y Asher, Nicholas
%Y Kim, Seokhwan
%Y Merlin, Teva
%S Proceedings of the 26th Annual Meeting of the Special Interest Group on Discourse and Dialogue
%D 2025
%8 August
%I Association for Computational Linguistics
%C Avignon, France
%F vlachos-etal-2025-building
%X We consider open-retrieval conversational question answering (OR-CONVQA), an extension of question answering where system responses need to be (i) aware of dialog history and (ii) grounded in documents (or document fragments) retrieved per question. Domain-specific OR-CONVQA training datasets are crucial for real-world applications, but hard to obtain. We propose a pipeline that capitalizes on the abundance of plain text documents in organizations (e.g., product documentation) to automatically produce realistic OR-CONVQA dialogs with annotations. Similarly to real-world humanannotated OR-CONVQA datasets, we generate in-dialog question-answer pairs, self-contained (decontextualized, e.g., no referring expressions) versions of user questions, and propositions (sentences expressing prominent information from the documents) the system responses are grounded in. We show how the synthetic dialogs can be used to train efficient question rewriters that decontextualize user questions, allowing existing dialog-unaware retrievers to be utilized. The retrieved information and the decontextualized question are then passed on to an LLM that generates the system’s response.
%U https://aclanthology.org/2025.sigdial-1.52/
%P 641-657
Markdown (Informal)
[Building Open-Retrieval Conversational Question Answering Systems by Generating Synthetic Data and Decontextualizing User Questions](https://aclanthology.org/2025.sigdial-1.52/) (Vlachos et al., SIGDIAL 2025)
ACL