EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation

Shaurya Gupta; Jatin Bedi

EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation

Correct Metadata for

Use this form to create a GitHub issue with structured data describing the correction. You will need a GitHub account. Once you create that issue, the correction will be reviewed by a staff member.

⚠️ Mobile Users: Submitting this form to create a new issue will only work with github.com, not the GitHub Mobile app.

Important: The Anthology treat PDFs as authoritative. Please use this form only to correct data that is out of line with the PDF. See our corrections guidelines if you need to change the PDF.

Title Adjust the title. Retain tags such as <fixed-case>.

Authors Adjust author names and order to match the PDF.

Abstract Correct abstract if needed. Retain XML formatting tags such as <tex-math>. You may use <b>...</b> for bold, <i>...</i> for italic, and <url>...</url> for URLs.

Verification against PDF Ensure that the new title/authors match the snapshot below. (If there is no snapshot or it is too small, consult the PDF.)

Authors concatenated from the text boxes above:

ALL author names match the snapshot above—including middle initials, hyphens, and accents.

Abstract

We describe EFSG (Evidence-First Structured Generation), our submission to Task B of the RAG4Reports@ACL 2026 shared task. Standard retrieval-augmented generation pipelines allow generation models to write from parametric memory and attach citations retroactively: a behaviour we term post-rationalization. EFSG addresses this structurally through a phase boundary: all evidence is retrieved, extracted, and sealed into a fact pool before any generation begins; each sentence then sees only its single committed source passage. Our best run (t5100k doc corpus) achieved sentence_support of 0.612 and nugget_coverage of 0.126 (F1 = 0.182).

Anthology ID:: 2026.rag4reports-1.14
Volume:: Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026)
Month:: July
Year:: 2026
Address:: San Diego, CA, USA
Editors:: Eugene Yang, Dawn Lawrie, Sean MacAvaney, James Mayfield, Luca Soldaini, Andrew Yates
Venues:: RAG4Reports | WS
SIG:
Publisher:: Association for Computational Linguistics
Note:
Pages:: 99–102
Language:
URL:: https://aclanthology.org/2026.rag4reports-1.14/
DOI:
Bibkey:
Cite (ACL):: Shaurya Gupta and Jatin Bedi. 2026. EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation. In Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026), pages 99–102, San Diego, CA, USA. Association for Computational Linguistics.
Cite (Informal):: EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation (Gupta & Bedi, RAG4Reports 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.rag4reports-1.14.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{gupta-bedi-2026-efsg,
    title = "{EFSG}: Evidence-First Structured Generation for Multilingual {RAG} Report Generation",
    author = "Gupta, Shaurya  and
      Bedi, Jatin",
    editor = "Yang, Eugene  and
      Lawrie, Dawn  and
      MacAvaney, Sean  and
      Mayfield, James  and
      Soldaini, Luca  and
      Yates, Andrew",
    booktitle = "Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation ({RAG}4{R}eports 2026)",
    month = jul,
    year = "2026",
    address = "San Diego, CA, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.rag4reports-1.14/",
    pages = "99--102",
    ISBN = "979-8-89176-417-0",
    abstract = "We describe EFSG (Evidence-First Structured Generation), our submission to Task B of the RAG4Reports@ACL 2026 shared task. Standard retrieval-augmented generation pipelines allow generation models to write from parametric memory and attach citations retroactively: a behaviour we term post-rationalization. EFSG addresses this structurally through a phase boundary: all evidence is retrieved, extracted, and sealed into a fact pool before any generation begins; each sentence then sees only its single committed source passage. Our best run (t5100k doc corpus) achieved sentence{\_}support of 0.612 and nugget{\_}coverage of 0.126 (F1 = 0.182)."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gupta-bedi-2026-efsg">
    <titleInfo>
        <title>EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Shaurya</namePart>
        <namePart type="family">Gupta</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jatin</namePart>
        <namePart type="family">Bedi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Eugene</namePart>
            <namePart type="family">Yang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Dawn</namePart>
            <namePart type="family">Lawrie</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sean</namePart>
            <namePart type="family">MacAvaney</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">James</namePart>
            <namePart type="family">Mayfield</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Luca</namePart>
            <namePart type="family">Soldaini</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Andrew</namePart>
            <namePart type="family">Yates</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego, CA, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-417-0</identifier>
    </relatedItem>
    <abstract>We describe EFSG (Evidence-First Structured Generation), our submission to Task B of the RAG4Reports@ACL 2026 shared task. Standard retrieval-augmented generation pipelines allow generation models to write from parametric memory and attach citations retroactively: a behaviour we term post-rationalization. EFSG addresses this structurally through a phase boundary: all evidence is retrieved, extracted, and sealed into a fact pool before any generation begins; each sentence then sees only its single committed source passage. Our best run (t5100k doc corpus) achieved sentence_support of 0.612 and nugget_coverage of 0.126 (F1 = 0.182).</abstract>
    <identifier type="citekey">gupta-bedi-2026-efsg</identifier>
    <location>
        <url>https://aclanthology.org/2026.rag4reports-1.14/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>99</start>
            <end>102</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation
%A Gupta, Shaurya
%A Bedi, Jatin
%Y Yang, Eugene
%Y Lawrie, Dawn
%Y MacAvaney, Sean
%Y Mayfield, James
%Y Soldaini, Luca
%Y Yates, Andrew
%S Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, CA, USA
%@ 979-8-89176-417-0
%F gupta-bedi-2026-efsg
%X We describe EFSG (Evidence-First Structured Generation), our submission to Task B of the RAG4Reports@ACL 2026 shared task. Standard retrieval-augmented generation pipelines allow generation models to write from parametric memory and attach citations retroactively: a behaviour we term post-rationalization. EFSG addresses this structurally through a phase boundary: all evidence is retrieved, extracted, and sealed into a fact pool before any generation begins; each sentence then sees only its single committed source passage. Our best run (t5100k doc corpus) achieved sentence_support of 0.612 and nugget_coverage of 0.126 (F1 = 0.182).
%U https://aclanthology.org/2026.rag4reports-1.14/
%P 99-102

Download as File

Markdown (Informal)

[EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation](https://aclanthology.org/2026.rag4reports-1.14/) (Gupta & Bedi, RAG4Reports 2026)

EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation (Gupta & Bedi, RAG4Reports 2026)

ACL

Shaurya Gupta and Jatin Bedi. 2026. EFSG: Evidence-First Structured Generation for Multilingual RAG Report Generation. In Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026), pages 99–102, San Diego, CA, USA. Association for Computational Linguistics.