Graph Topology Meets Energy-Driven Pathfinding: Preserving Contextual Consistency in Retrieval-Augmented Generation

Yufeng Li; Zaijing Li; Fuchen Zhang; Zongzong Wu; Ming Zhao; Fengxiao Tang

cs.CLOct 21, 2026

Graph Topology Meets Energy-Driven Pathfinding: Preserving Contextual Consistency in Retrieval-Augmented Generation

Authors: Yufeng Li, Zaijing Li, Fuchen Zhang, Zongzong Wu, Ming Zhao, Fengxiao Tang

Organizations: Central South University, Changsha, China · Harbin Institute of Technology, Shenzhen, China

Abstract

Retrieval-augmented generation (RAG) frequently integrates knowledge graphs (KGs) for complex multi-hop question answering. However, their performance is increasingly bottlenecked by the knowledge structuring and reasoning processes. (i) Severe semantic loss: Distilling continuous texts into discrete symbols strips away contextual nuances and narrative coherence. (ii) Pronounced modality mismatch: Fragmented graph structures clash with the continuous texts LLMs are pre-trained on, while rewriting them into natural language introduces generative noise and omits information. To address these limitations, we propose ThreadRAG, a semantic-topology-based RAG framework for context-preserving knowledge structuring and modality-native evidence reasoning. ThreadRAG introduces a Semantic Topology Mapper (STM), which organizes retrieved documents into context-intact Semantic Nodes and constructs a Semantic Topology Graph (STG) through narrative, semantic, and lexical edges. It further incorporates Energy-Driven Pathfinding (EDP), which identifies query-relevant subgraphs via energy diffusion and deterministically linearizes them into coherent textual evidence paths. Experiments on three multi-hop QA benchmarks demonstrate that ThreadRAG consistently outperforms strong baselines. Notably, it achieves a 14.4% EM improvement over text-centric methods on HotpotQA and surpasses the best baseline by 5.8% EM on 2WikiMultihop, while reducing token cost to 23.3% of the strongest baseline.

Figures & tables

Figure 1: Comparison of graph-RAG paradigms. Left : Our proposed ThreadRAG (a) structures raw texts into a Semantic Topology Graph (STG) and extracts Semantic Topology Paths (ST-Paths), preserving semantic context and aligning modality for LLM deduction. Right : Existing paradigms, Triplet KG (b) and Chunk KG (c), fragment texts into discrete components, introducing severe semantic loss and modality mismatch.

Figure 2: Overview of ThreadRAG. Given a query and top- $k$ retrieved contexts, the STM structures raw text into STG. The EDP then isolates a query-relevant subgraph via energy diffusion and linearizes it into modality-native prompts. Finally, an adaptive agent uses these prompts to iteratively refine retrieval or synthesize answers.

Methods	LLaMA3-8B-Instruct						Qwen3.5 plus						GPT-4o mini
	HotpotQA		2Wiki		MuSiQue		HotpotQA		2Wiki		MuSiQue		HotpotQA		2Wiki		MuSiQue
	EM	F1	EM	F1	EM	F1	EM	F1	EM	F1	EM	F1	EM	F1	EM	F1	EM	F1
Non-Graph RAG
RA-CoT	.204	.330	.202	.261	.142	.210	.312	.429	.268	.372	.162	.249	.310	.425	.264	.359	.158	.243
ReAct	.268	.379	.244	.335	.216	.311	.358	.476	.342	.487	.248	.378	.349	.475	.338	.483	.242	.383
IRCoT	.318	.430	.288	.367	.224	.327	.442	.565	.374	.519	.262	.376	.438	.559	.372	.496	.258	.373

Table 1: Main results of multi-hop QA datasets. Best results are in bold and second best are underlined .

Ablation Setting			Datasets
SN	STG	EDP	HotpotQA	2Wiki	MuSiQue
✓	✓	✓	.422	.372	.324
			.238 ( $\downarrow$ 18.4%)	.224 ( $\downarrow$ 14.8%)	.172 ( $\downarrow$ 15.2%)
		✓	.326 ( $\downarrow$ 9.6%)	.308 ( $\downarrow$ 6.4%)	.232 ( $\downarrow$ 9.2%)
✓	✓		.398 ( $\downarrow$ 2.4%)	.334 ( $\downarrow$ 3.8%)	.286 ( $\downarrow$ 3.8%)
	✓	✓	.388 ( $\downarrow$ 3.4%)	.342 ( $\downarrow$ 3%)	.292 ( $\downarrow$ 3.2%)

Table 2: Ablation study of the STM (SN, STG) and EDP modules across three datasets.

Figure 3: Hyperparameter sensitivity analysis on the HotpotQA dataset.

Method	HotpotQA			MuSiQue
Method	Const.	Inf.	Total	Const.	Inf.	Total
LightRAG	26.2M	47.4k	26.25M	36.4M	57.6k	36.46M
KnowTrace	0.15M	72.6k	222.6k	0.22M	94.4k	314.4k
ThreadRAG	0	58.5k	58.5k	0	73.4k	73.4k

Table 3: Comparison of construction (Const.), inference (Inf.), and total token costs.

Figure 4: Performance comparison of EM scores using different underlying graph representations.

Figure 5: Performance comparison of various knowledge representation strategies.

Appendix figures & tables8 assets

Supplementary material from the paper’s appendix.

Appendix

Question: Who sings home alone tonight with the singer of you can crash my party anytime?

[Phase 1: Query Analysis & Retrieval]

# Entity Extraction:

- Entities to retrieve: "Home Alone Tonight", "Crash My Party"

# Retrieve Relevant Textual Passages:

- Take Me Home Tonight (song): "Take Me Home Tonight" is a song by American rock singer Eddie Money…

- Luke Bryan discography: American country music singer and songwriter Luke Bryan has released…

Appendix

Table 4: Case on MuSiQue.

Prompt 3: Dataset Generation

You are an expert linguistic annotator. Your task is to perform “Semantic Node Partitioning” on a given multi-hop document. Your goal is to divide the continuous text into self-contained “Semantic Nodes” along natural narrative, logical, or topical breaks.

## Rules for Partitioning:

1. Contextual Integrity: Each semantic node should be a coherent unit of information (e.g., a complete description of an entity, a specific event, or a logical step). Do not split sentences that are tightly coupled (e.g., pronouns referring to the immediately preceding sentence).

2. Granularity: A node typically consists of 1 to 2 sentences. Avoid creating overly long nodes that merge distinct topics, and avoid overly granular nodes (e.g., splitting a single compound sentence).

3. Exact Match: The text of the partitioned nodes MUST perfectly reconstruct the original document when concatenated. Do not alter, add, or remove any words.

## Output Format:

Appendix

Table 5: Prompt Templates of Dataset Generation.

Method	HotpotQA	2Wiki	MuSiQue
Graph-Free	0.274	0.258	0.206
ThreadRAG	0.422	0.372	0.324

Appendix

Table 6: Controlled Graph-Free ablation in terms of Exact Match (EM). The Graph-Free variant preserves retrieval and iterative reasoning while removing graph construction and EDP.

Subgraph Strategy	EM	$\boldsymbol{\Delta}$ vs. EDP
No subgraph	0.286	$-0.038$
BFS (depth = 3)	0.294	$-0.030$
DFS (depth = 3)	0.302	$-0.022$
EDP	0.324	–

Appendix

Table 7: Comparison of evidence selection strategies on MuSiQue using LLaMA3-8B-Instruct. All methods use the same candidate graph and evidence budget.

Precision	Recall	F1
0.948	0.861	0.902

Appendix

Table 8: Intrinsic evaluation of the semantic boundary predictor on 50 held-out Wikipedia documents. Scores measure agreement with teacher-generated reference boundaries.

Prompt 1: Entity Extraction

You are an entity extraction assistant. Extract key entities from the question for knowledge retrieval.

Rules:

1. Extract only key entities (persons, places, organizations, products, etc.)

2. Return ONLY a JSON array of entity strings

3. DO NOT include generic words like "who", "what", "where", "the", "a", etc.

4. For multi-hop questions, include ALL entities needed for reasoning

Appendix

Table 9: Prompt Templates of Entity Extraction.

Prompt 2: Discriminator

You are an Information Sufficiency Discriminator.

Your task is to evaluate whether the provided context is sufficient to answer the user’s question.

## Output Format (JSON):

{"status": "Sufficient" or "Insufficient", "reason": "brief explanation if insufficient, empty string if sufficient"}

## Evaluation Criteria:

- If the context contains enough factual information to answer the question, output {"status": "Sufficient", "reason": ""}

Appendix

Table 10: Prompt Templates of Discriminator.

Figure 6: Distribution of failure modes analyzed from 200 incorrect predictions.

Graph Retrieval-Augmented Generation

Graph Retrieval-Augmented Generation connects language models to knowledge graphs to retrieve relational evidence for grounded, multi-hop answers.

See all 232 papers

Efficient Retrieval-Augmented Generation via Token Co-occurrence Graphs
Jun 29, 2026Gianluca Bonifazi, Christopher Buratti, Michele Marchetti +5Multi-Hop QA Graph Retrieval-Augmented Generation
aUniversità Politecnica delle Marche, Ancona, Italy · bUniversità di Modena e Reggio Emilia, Modena, Italy
SAG: SQL-Retrieval Augmented Generation with Query-Time Dynamic Hyperedges
Aug 12, 2026Yuchao Wu, Junqin Li, XingCheng Liang +4Multi-Hop QA Retrieval-Augmented Generation
Zleap AI
The Reasoning Bottleneck in Graph-RAG: Structured Prompting and Context Compression for Multi-Hop QA
Mar 14, 2026Yasaman Zarrinkia, Venkatesh Srinivasan, Alex ThomoStructured CoT Reasoning Graph Retrieval-Augmented Generation
University of Victoria · Santa Clara University

Jun 29, 2026J/K move · Enter open · S save

Efficient Retrieval-Augmented Generation via Token Co-occurrence Graphs

Gianluca Bonifazi, Christopher Buratti, Michele Marchetti, Federica Parlapiano+4

aUniversità Politecnica delle Marche, Ancona, Italy · bUniversità di Modena e Reggio Emilia, Modena, Italy

Multi-Hop QA Graph Retrieval-Augmented Generation Graph-Based Retrieval

Open paper AskPDF arXiv

Retrieval-Augmented Generation (RAG) mitigates hallucinations in Large Language Models (LLMs) by grounding the generation process on external knowledge. However, standard RAG approaches struggle with multi-hop reasoning. While recent graph-based RAG methods improve the retrieval of interconnected chunks, they often rely on computationally expensive and error-prone LLM-based extraction pipelines. To address these issues, we propose TIGRAG (Token-Induced GraphRAG), an efficient graph-augmented RAG framework based on a token co-occurrence Knowledge Graph. TIGRAG directly models topological relationships between tokens using sliding-window co-occurrence statistics, thus enabling scalable graph construction. During inference, it combines graph-based semantic expansion and neural reranking to retrieve interconnected evidence for multi-hop reasoning. Specifically, it introduces an iterative entity-driven retrieval strategy that progressively expands the query using bridging entities extracted from previously retrieved contexts. We evaluated TIGRAG on three widely adopted multi-hop Question Answering (QA) benchmarks. Experimental results demonstrated that our framework consistently outperforms dense retrieval and graph-based RAG methods in both retrieval and downstream QA tasks, while substantially reducing indexing time, inference latency, and prompt footprint.

Abstract

Graph Retrieval-Augmented Generation

Explore similar work

Efficient Retrieval-Augmented Generation via Token Co-occurrence Graphs

SAG: SQL-Retrieval Augmented Generation with Query-Time Dynamic Hyperedges

The Reasoning Bottleneck in Graph-RAG: Structured Prompting and Context Compression for Multi-Hop QA