# RFKG-CoT: Relation-Driven Adaptive Hop-count Selection and Few-Shot Path Guidance for Knowledge-Aware QA

Source: [RFKG-CoT: Relation-Driven Adaptive Hop-count Selection and Few-Shot Path Guidance for Knowledge-Aware QA](<https://www.sota2.com/research/paper/69438a74423ea422657293a2>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-19T05:00:04.527Z

Authors: Chao Zhang, Minghan Li, Tianrui Lv, Guodong Zhou

Published: 2025-12-17T09:14:08.000Z

## Abstract

Large language models (LLMs) often generate hallucinations in knowledge-intensive QA due to parametric knowledge limitations. While existing methods like KG-CoT improve reliability by integrating knowledge graph (KG) paths, they suffer from rigid hop-count selection (solely question-driven) and underutilization of reasoning paths (lack of guidance). To address this, we propose RFKG-CoT: First, it replaces the rigid hop-count selector with a relation-driven adaptive hop-count selector that dynamically adjusts reasoning steps by activating KG relations (e.g., 1-hop for direct "brother" relations, 2-hop for indirect "father-son" chains), formalized via a relation mask. Second, it introduces a few-shot in-context learning path guidance mechanism with CoT (think) that constructs examples in a "question-paths-answer" format to enhance LLMs' ability to understand reasoning paths. Experiments on four KGQA benchmarks show RFKG-CoT improves accuracy by up to 14.7 pp (Llama2-7B on WebQSP) over KG-CoT. Ablations confirm the hop-count selector and the path prompt are complementary, jointly transforming KG evidence into more faithful answers.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Multi-hop QA on WebQSP](<https://www.sota2.com/research/sota/multi-hop-qa-on-webqsp.md>): Multi-hop QA; WebQSP; Accuracy: 91.5.

- [Single-hop QA on SimpleQuestions](<https://www.sota2.com/research/sota/single-hop-qa-on-simplequestions.md>): Single-hop QA; SimpleQuestions; Accuracy: 0.87.

- [Open-domain QA on WebQuestions](<https://www.sota2.com/research/sota/open-domain-qa-on-webquestions.md>): Open-domain QA; WebQuestions; Accuracy: 78.2.

- [Multi-hop QA on CompWebQ](<https://www.sota2.com/research/sota/multi-hop-qa-on-compwebq.md>): Multi-hop QA; CompWebQ; Accuracy: 65.1.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.15219>)

[PDF](<https://arxiv.org/pdf/2512.15219.pdf>)