Saved in:
Bibliographic Details
Main Authors: Wang, Chenyu, Liu, Yingmin, Shu, Yang
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.17301
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910228688142336
author Wang, Chenyu
Liu, Yingmin
Shu, Yang
author_facet Wang, Chenyu
Liu, Yingmin
Shu, Yang
contents Retrieval-Augmented Generation (RAG) systems implicitly assume mutual consistency among retrieved documents -- an assumption that frequently fails in practice. We present ConflictRAG, a conflict-aware RAG framework that detects, classifies, and resolves knowledge conflicts prior to answer generation. The framework introduces three contributions: (1) a two-stage conflict detection module combining a lightweight embedding-based MLP classifier with selective LLM refinement, reducing API costs by 62% while maintaining 90.8% detection accuracy; (2) an Entropy-TOPSIS framework for data-driven source credibility assessment, improving selection accuracy by 7.1% over manual heuristics; and (3) a Conflict-Aware RAG Score (CARS) for diagnostic evaluation of conflict-handling capabilities. Experiments on three benchmarks against six baselines demonstrate 88.7% conflict-detection F1 and consistent 5.3--6.1% correctness gains over the strongest conflict-aware baseline, with the pipeline transferring effectively across backbone LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2605_17301
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ConflictRAG: Detecting and Resolving Knowledge Conflicts in Retrieval Augmented Generation
Wang, Chenyu
Liu, Yingmin
Shu, Yang
Computation and Language
Artificial Intelligence
Retrieval-Augmented Generation (RAG) systems implicitly assume mutual consistency among retrieved documents -- an assumption that frequently fails in practice. We present ConflictRAG, a conflict-aware RAG framework that detects, classifies, and resolves knowledge conflicts prior to answer generation. The framework introduces three contributions: (1) a two-stage conflict detection module combining a lightweight embedding-based MLP classifier with selective LLM refinement, reducing API costs by 62% while maintaining 90.8% detection accuracy; (2) an Entropy-TOPSIS framework for data-driven source credibility assessment, improving selection accuracy by 7.1% over manual heuristics; and (3) a Conflict-Aware RAG Score (CARS) for diagnostic evaluation of conflict-handling capabilities. Experiments on three benchmarks against six baselines demonstrate 88.7% conflict-detection F1 and consistent 5.3--6.1% correctness gains over the strongest conflict-aware baseline, with the pipeline transferring effectively across backbone LLMs.
title ConflictRAG: Detecting and Resolving Knowledge Conflicts in Retrieval Augmented Generation
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2605.17301