@inproceedings{dhurve-etal-2026-decompose,
title = "Decompose, Retrieve, Cite: A {RAG} Pipeline for Structured Report Generation from Technical Documentation",
author = "Dhurve, Himanshu and
Panat, Sreedath and
Dandekar, Rajat and
Dandekar, Raj",
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-1.4/",
pages = "24--35",
ISBN = "979-8-89176-417-0",
abstract = "Retrieval-Augmented Generation (RAG) grounds language-model output in external knowledge, yet its application to dense technical documentation remains largely unexplored. Engineering software manuals pose compounding challenges: formulae are corrupted during PDF extraction, heterogeneous content types require different parsing treatment, and queries demand cross-document synthesis across multiple reference volumes.We present an end-to-end RAG system for OpenFOAM, an open-source computational fluid dynamics toolkit, operating in two modes. In single-query mode, a formula-preserving parser (Marker), adaptive header-aware chunking, two-stage dense-then-rerank retrieval, and a citation-enforcement prompt produce grounded, source-attributed answers across a 20-question benchmark.In report mode, a user prompt is decomposed into sub-questions via LLM planning; each sub-question undergoes independent retrieval and cross-encoder re-ranking, and the deduplicated chunk set is passed to a long-context generation call that produces a structured, multi-section report with inline citations.Evaluated on a 10-prompt golden set with a six-dimension LLM-as-a-judge framework, both pipelines achieve overall scores above 4.6/5.0 with perfect citation correctness (5.0/5.0). The decomposed pipeline demonstrates superior robustness (90{\%} vs 70{\%} judge success rate). Retrieval analysis using page-level ground truth reveals low absolute recall ({\ensuremath{<}}14{\%}), identifying retrieval breadth as the primary bottleneck."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dhurve-etal-2026-decompose">
<titleInfo>
<title>Decompose, Retrieve, Cite: A RAG Pipeline for Structured Report Generation from Technical Documentation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Himanshu</namePart>
<namePart type="family">Dhurve</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sreedath</namePart>
<namePart type="family">Panat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rajat</namePart>
<namePart type="family">Dandekar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raj</namePart>
<namePart type="family">Dandekar</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>Retrieval-Augmented Generation (RAG) grounds language-model output in external knowledge, yet its application to dense technical documentation remains largely unexplored. Engineering software manuals pose compounding challenges: formulae are corrupted during PDF extraction, heterogeneous content types require different parsing treatment, and queries demand cross-document synthesis across multiple reference volumes.We present an end-to-end RAG system for OpenFOAM, an open-source computational fluid dynamics toolkit, operating in two modes. In single-query mode, a formula-preserving parser (Marker), adaptive header-aware chunking, two-stage dense-then-rerank retrieval, and a citation-enforcement prompt produce grounded, source-attributed answers across a 20-question benchmark.In report mode, a user prompt is decomposed into sub-questions via LLM planning; each sub-question undergoes independent retrieval and cross-encoder re-ranking, and the deduplicated chunk set is passed to a long-context generation call that produces a structured, multi-section report with inline citations.Evaluated on a 10-prompt golden set with a six-dimension LLM-as-a-judge framework, both pipelines achieve overall scores above 4.6/5.0 with perfect citation correctness (5.0/5.0). The decomposed pipeline demonstrates superior robustness (90% vs 70% judge success rate). Retrieval analysis using page-level ground truth reveals low absolute recall (\ensuremath<14%), identifying retrieval breadth as the primary bottleneck.</abstract>
<identifier type="citekey">dhurve-etal-2026-decompose</identifier>
<location>
<url>https://aclanthology.org/2026.rag4reports-1.4/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>24</start>
<end>35</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Decompose, Retrieve, Cite: A RAG Pipeline for Structured Report Generation from Technical Documentation
%A Dhurve, Himanshu
%A Panat, Sreedath
%A Dandekar, Rajat
%A Dandekar, Raj
%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 dhurve-etal-2026-decompose
%X Retrieval-Augmented Generation (RAG) grounds language-model output in external knowledge, yet its application to dense technical documentation remains largely unexplored. Engineering software manuals pose compounding challenges: formulae are corrupted during PDF extraction, heterogeneous content types require different parsing treatment, and queries demand cross-document synthesis across multiple reference volumes.We present an end-to-end RAG system for OpenFOAM, an open-source computational fluid dynamics toolkit, operating in two modes. In single-query mode, a formula-preserving parser (Marker), adaptive header-aware chunking, two-stage dense-then-rerank retrieval, and a citation-enforcement prompt produce grounded, source-attributed answers across a 20-question benchmark.In report mode, a user prompt is decomposed into sub-questions via LLM planning; each sub-question undergoes independent retrieval and cross-encoder re-ranking, and the deduplicated chunk set is passed to a long-context generation call that produces a structured, multi-section report with inline citations.Evaluated on a 10-prompt golden set with a six-dimension LLM-as-a-judge framework, both pipelines achieve overall scores above 4.6/5.0 with perfect citation correctness (5.0/5.0). The decomposed pipeline demonstrates superior robustness (90% vs 70% judge success rate). Retrieval analysis using page-level ground truth reveals low absolute recall (\ensuremath<14%), identifying retrieval breadth as the primary bottleneck.
%U https://aclanthology.org/2026.rag4reports-1.4/
%P 24-35
Markdown (Informal)
[Decompose, Retrieve, Cite: A RAG Pipeline for Structured Report Generation from Technical Documentation](https://aclanthology.org/2026.rag4reports-1.4/) (Dhurve et al., RAG4Reports 2026)
ACL