PRIME: Pre-Retrieval Index Modeling Enhancement for Structured Document Retrieval-Real

Cheuk Hei Chung; Li Liu

cs.IROct 9, 2026

PRIME: Pre-Retrieval Index Modeling Enhancement for Structured Document Retrieval-Real

Authors: Cheuk Hei Chung, Li Liu

Organizations: The Hong Kong University of Science and Technology (Guangzhou)

Abstract

High-precision retrieval over external knowledge bases is essential for LLM agents to operate effectively. However, existing Information Retrieval (IR) research primarily optimizes query-side retrievers (e.g., ColBERT, BGE) and generators (e.g., Self-RAG, GraphRAG), while reducing corpus-side representation to chunking. This leaves index-side quality an underexplored lever. We propose Pre-Retrieval Index Modeling & Enhancement (PRIME), an index-side optimization framework built on two complementary tracks: an LLM-driven Semantic Restructuring (SR) track that transforms long documents into structured, information-dense sections, and a deterministic Metadata Synthesis (MS) and Hierarchical Retrieval (HR) track that extracts semantic anchors and heading structure for coarse-to-fine section retrieval. Our evaluations show that deterministic metadata matches structure-aware chunking's retrieval quality while cutting embedding computation by 45.8 $\times$ and index vector count by 22.4 $\times$ . On top of these gains, Semantic Restructuring (SR) further improves Answer-F1 by 13.7% in full deployment, with characterized quality-fidelity-cost trade-offs. Across QASPER, BRIGHT, and NFCorpus, we further characterize three qualitative regimes that determine whether index-side optimization helps or degrades retrieval, while showing that these index-side effects are neutralized under a strong reranker.

Figures & tables

Figure 1: Overview of Index-Side Optimization. PRIME operates on the corpus representation prior to indexing, complementing existing query-side and generator-side methods without modifying the retriever.

Figure 2: PRIME architecture. The framework operates on two complementary tracks: (Left) an LLM-driven Semantic Restructuring track transforms documents into knowledge-dense representations; (Right) a deterministic track uses Metadata Synthesis to extract structural anchors, which power a Hierarchical Retrieval pipeline with RRF fusion and cross-encoder reranking.

Method	NDCG5	NDCG10	Recall5	Precision5	P1	MRR
Raw (chunk-256)	0.391	0.405	0.462	0.092	0.306	0.373
BGE-M3 (1024-dim)	0.389	0.398	0.453	0.091	0.306	0.371
Summary-Augmented	0.398	0.411	0.471	0.094	0.309	0.379
RAPTOR-lite	0.391	0.408	0.461	0.092	0.307	0.375
PRIME (offline) (ours)	0. 396	0.409	0.469	0.094	0.309	0.377
Structure-chunks	0.396	0.411	0.466	0.093	0.310	0.379

Table 1: Retrieval quality on QASPER 416 (1,451 queries). BGE-M3 uses chunk-256 max-pooling; Summary-Augmented prepends section summaries to chunks; RAPTOR-lite does section+window retrieval without LLM summarization. Structure-chunks is a simpler H2-aware chunking baseline (no MS metadata, no HR pipeline). PRIME (Full) is excluded because SR alters the BM25-indexed text, breaking the shared-pipeline condition.

Index Representation	Dense-only	Hybrid (BM25+Vec+RRF)	Hybrid+CE
Raw (chunk-256)	0.175 [.164,.186]	0.311 [.296,.327]	0.391 [.370,.412]
Structure-chunks	0.201 [.187,.216]	0.338 [.321,.354]	0.396 [.375,.416]
PRIME (offline)	0.201 [.186,.217]	0.332 [.315,.349]	0.396 [.375,.416]
SR (full-text)	0.208 [.196,.221]	0.245 [.232,.258]	0.328 [.309,.347]
PRIME (Full)	0.214 [.199,.229]	0.241 [.227,.256]	0.325 [.307,.345]

Table 2: Factorial ablation: NDCG@5 on QASPER 416. 95% paired bootstrap CIs in brackets. All configurations share identical retrieval pipelines; only the indexed text varies between rows. The five index representations and three retrieval pipelines are defined in Preliminary.

Tasks	BGE-small (baseline)			PRIME (offline)			$\Delta$ NDCG@5	95% CI ( $\Delta$ )
Tasks	NDCG@5	P@5	R@5	NDCG@5	P@5	R@5	$\Delta$ NDCG@5	95% CI ( $\Delta$ )
Pony	0.067	0.073	0.048	0.185	0.154	0.126	$+175.6\%$	$[-0.063,\ +0.299]$
Biology	0.379	0.128	0.491	0.464	0.155	0.598	$+22.3\%$	$[-0.188,\ +0.357]$
Stackoverflow	0.234	0.075	0.359	0.272	0.079	0.363	$+16.2\%$	$[-0.203,\ +0.279]$
Robotics	0.181	0.052	0.253	0.200	0.059	0.282	$+10.5\%$	$[-0.250,\ +0.290]$
Economics	0.247	0.070	0.333	0.258	0.074	0.354	$+4.6\%$	$[-0.242,\ +0.265]$

Table 3: Cross-domain retrieval on the eight BRIGHT structured tasks.

\Delta

= (PRIME-Baseline)/Baseline; 95% CIs from 10,000-sample paired bootstrap (BCa) over per-query NDCG@5. Per-task CIs span zero (65-103 queries/task), so individual task deltas are not independently significant; the +6.9% macro-average reflects a distributional shift across domains rather than a per-task claim. TheoremQA-Theorems (PRIME NDCG@5 = 0.012,

\Delta

= -59.8%) is excluded from this average.

Method	NDCG@5	P@5	R@5	MRR
BM25	0.350	0.295	0.119	0.530
BGE-small	0.374	0.324	0.121	0.538
BGE-base	0.387	0.336	0.126	0.561
PRIME	0.301	0.264	0.084	0.460
PRIME (FT)	0.366	0.316	0.118	0.540
$\Delta_{1}$	-19.5%	-18.5%	-30.6%	-14.5%

Table 4: Retrieval quality on the full NFCorpus. Graded relevance 0–2 binarized at

\geq 1

. PRIME(FT) denotes PRIME (Full-Text),

\Delta_{1}

denotes PRIME vs BGE-small and

\Delta_{2}

denotes PRIME (Full-Text) vs BGE-small.

Panel A: Corpus-level density (416 docs)
	Raw	SR	$\Delta$ %
Stopword ratio	0.365	0.233	-36.3%
Mean IDF	1.947	2.331	+19.7%
Tech term density (per 100 toks)	46.3	53.8	+16.1%
Semantic density (composite)	3.42	7.76	+127.3%
Panel B: Section-level evidence retrieval (CE reranker)

Table 5: SR causal chain: text density and evidence localization (QASPER 416). Panel A uses fixed raw-corpus reference IDF. Panel B uses section qrels from gold evidence spans mapped to H2 sections, averaged over the 708 shared queries of Raw and SR section qrels to avoid survivor bias.

Metrics	Value
Panel A: Evidence retention (2,231 gold spans, 373 docs)
Overall span retention	63.2%
Median per-document retention	66.7% (p25: 44.4%, p75: 100%)
Perfect retention (100% of spans)	26.8% of docs
High retention (80–99%)	35.1% of docs
Low retention (<50%)	25.7% of docs

Table 6: SR fidelity audit on QASPER. Panel A: a span is retained if its character-level content appears verbatim in the SR block. Panel B uses spaCy en_core_web_sm. Panel C uses roberta-large-mnli.

Information Retrieval

Information Retrieval studies how to find, rank, and organize relevant information in large collections of data.

See all 474 papers

RIT-RAG: Navigating Document Corpora with Retrieval-Induced Trees
Oct 8, 2026Meghanadh Pulivarthi, Swaraj Kumar Biswal, Kushagra Bhushan +3Retrieval-Augmented Generation Structured RAG
IBM · IIT Kharagpur
MCompassRAG: Topic Metadata as a Semantic Compass for Paragraph-Level Retrieval
Jun 16, 2026Amirhossein Abaskohi, Raymond Li, Gaetano Cimino +3Retrieval-Augmented Generation Information Retrieval
University of British Columbia · University of Salerno · ServiceNow Research
RL-Index: Reinforcement Learning for Retrieval Index Reasoning
Jun 15, 2026Yongjia Lei, Nedim Lipka, Zhisheng Qi +5Reinforcement Learning Agentic Retrieval
♡University of Oregon · ♣Adobe Research

Oct 8, 2026J/K move · Enter open · S save

RIT-RAG: Navigating Document Corpora with Retrieval-Induced Trees

Meghanadh Pulivarthi, Swaraj Kumar Biswal, Kushagra Bhushan, Yatin Nandwani+2

IBM · IIT Kharagpur

Retrieval-Augmented Generation Structured RAG Agentic RAG Question Answering

Open paper AskPDF arXiv

Retrieval-augmented generation (RAG) grounds language models in external corpora. Agentic RAG enables iterative search, yet exposes the model to isolated chunks without document structure, making it difficult to distinguish relevant evidence from chunks that merely resemble the query. Structure-aware methods such as PageIndex navigate document structure but cannot scale to the structures of large corpora, which do not fit in the LLM context. Hence, they first commit to a single document using a document retriever and cannot recover from a wrong choice. We propose RIT-RAG (Retrieval-Induced Tree RAG), which combines content retrieval with structural navigation. Offline, RIT-RAG builds a tree for each document from its table of contents or sitemap. At query time, it retrieves a broad set of chunks and uses their positions to induce manageable sub-trees, potentially across multiple documents. An LLM agent navigates these sub-trees, selectively reads promising nodes, and reformulates queries when needed. Thus, retrieval proposes where to look, while the agent decides what to read. Across financial, scientific, and customer-support benchmarks, RIT-RAG achieves the highest answer accuracy among vanilla, graph-based, and agentic baselines. On EntQABench, our new benchmark of 2.84 million technical-documentation webpages, it improves accuracy by 6.8 to 11.4 points over the strongest baseline across three LLMs.

Abstract

Information Retrieval

Explore similar work

RIT-RAG: Navigating Document Corpora with Retrieval-Induced Trees

MCompassRAG: Topic Metadata as a Semantic Compass for Paragraph-Level Retrieval

RL-Index: Reinforcement Learning for Retrieval Index Reasoning