@inproceedings{ji-etal-2026-convergewriter,
title = "{C}onverge{W}riter: Data-Driven Bottom-Up Article Construction",
author = "Ji, Binquan and
Wang, Jiaqi and
Li, Ruiting and
Han, Xingchen and
Luo, Haibo and
Qi, Yiyang and
Wang, Shichao and
Lu, Yifei and
Han, Yuantao and
Ren, Feiliang",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.61/",
pages = "1238--1256",
ISBN = "979-8-89176-462-0",
abstract = "Large Language Models (LLMs) excel at text generation, but still struggle to produce accurate, long-form documents grounded in large knowledge bases. Existing ``top-down'' approaches{---}which plan before retrieving evidence{---}often result in fragmented or unfaithful content due to misalignment between the plan and available knowledge.\par To overcome this, we introduce a ``bottom-up'' framework based on a ``Retrieval-First for Knowledge, Clustering for Structure'' strategy. Our method begins by establishing the ``knowledge boundaries'' of the source corpus through comprehensive retrieval, followed by unsupervised clustering to group evidence into distinct ``knowledge clusters.'' These clusters objectively inform a hierarchical outline and subsequent text generation, ensuring strict adherence to the source and significantly reducing hallucination.\par Experimental results on both 14B- and 32B-parameter models demonstrate that our method achieves performance comparable to or exceeding state-of-the-art baselines, and is expected to demonstrate unique advantages in knowledge-constrained scenarios that demand high fidelity and structural coherence. Our work presents an effective paradigm for generating reliable, structured, long-form documents, paving the way for more robust LLM applications in high-stakes, knowledge-intensive domains."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ji-etal-2026-convergewriter">
<titleInfo>
<title>ConvergeWriter: Data-Driven Bottom-Up Article Construction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Binquan</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaqi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruiting</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xingchen</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haibo</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiyang</namePart>
<namePart type="family">Qi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shichao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yifei</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuantao</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Feiliang</namePart>
<namePart type="family">Ren</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) excel at text generation, but still struggle to produce accurate, long-form documents grounded in large knowledge bases. Existing “top-down” approaches—which plan before retrieving evidence—often result in fragmented or unfaithful content due to misalignment between the plan and available knowledge.\par To overcome this, we introduce a “bottom-up” framework based on a “Retrieval-First for Knowledge, Clustering for Structure” strategy. Our method begins by establishing the “knowledge boundaries” of the source corpus through comprehensive retrieval, followed by unsupervised clustering to group evidence into distinct “knowledge clusters.” These clusters objectively inform a hierarchical outline and subsequent text generation, ensuring strict adherence to the source and significantly reducing hallucination.\par Experimental results on both 14B- and 32B-parameter models demonstrate that our method achieves performance comparable to or exceeding state-of-the-art baselines, and is expected to demonstrate unique advantages in knowledge-constrained scenarios that demand high fidelity and structural coherence. Our work presents an effective paradigm for generating reliable, structured, long-form documents, paving the way for more robust LLM applications in high-stakes, knowledge-intensive domains.</abstract>
<identifier type="citekey">ji-etal-2026-convergewriter</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.61/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1238</start>
<end>1256</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ConvergeWriter: Data-Driven Bottom-Up Article Construction
%A Ji, Binquan
%A Wang, Jiaqi
%A Li, Ruiting
%A Han, Xingchen
%A Luo, Haibo
%A Qi, Yiyang
%A Wang, Shichao
%A Lu, Yifei
%A Han, Yuantao
%A Ren, Feiliang
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F ji-etal-2026-convergewriter
%X Large Language Models (LLMs) excel at text generation, but still struggle to produce accurate, long-form documents grounded in large knowledge bases. Existing “top-down” approaches—which plan before retrieving evidence—often result in fragmented or unfaithful content due to misalignment between the plan and available knowledge.\par To overcome this, we introduce a “bottom-up” framework based on a “Retrieval-First for Knowledge, Clustering for Structure” strategy. Our method begins by establishing the “knowledge boundaries” of the source corpus through comprehensive retrieval, followed by unsupervised clustering to group evidence into distinct “knowledge clusters.” These clusters objectively inform a hierarchical outline and subsequent text generation, ensuring strict adherence to the source and significantly reducing hallucination.\par Experimental results on both 14B- and 32B-parameter models demonstrate that our method achieves performance comparable to or exceeding state-of-the-art baselines, and is expected to demonstrate unique advantages in knowledge-constrained scenarios that demand high fidelity and structural coherence. Our work presents an effective paradigm for generating reliable, structured, long-form documents, paving the way for more robust LLM applications in high-stakes, knowledge-intensive domains.
%U https://aclanthology.org/2026.findings-emnlp.61/
%P 1238-1256
Markdown (Informal)
[ConvergeWriter: Data-Driven Bottom-Up Article Construction](https://aclanthology.org/2026.findings-emnlp.61/) (Ji et al., Findings 2026)
ACL
- Binquan Ji, Jiaqi Wang, Ruiting Li, Xingchen Han, Haibo Luo, Yiyang Qi, Shichao Wang, Yifei Lu, Yuantao Han, and Feiliang Ren. 2026. ConvergeWriter: Data-Driven Bottom-Up Article Construction. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 1238–1256, Budapest, Hungary. Association for Computational Linguistics.