AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

TruthfulRAG: Resolving Factual-level Conflicts in Retrieval-Augmented Generation with Knowledge Graphs

Shuyi Liu, Yu-Ming Shang, Xi Zhang

AAAI 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add knowledge-graph construction and retrieval to entropy filtering (ablation)TruthfulRAG (GPT-4o-mini) · FaithEvalaccuracy: 64.8 -> 69.5Improved
add knowledge-graph construction and retrieval to entropy filtering (ablation)TruthfulRAG (GPT-4o-mini) · SQuADaccuracy: 78.8 -> 81.1Improved
add knowledge-graph construction and retrieval to entropy filtering (ablation)TruthfulRAG (GPT-4o-mini) · RealtimeQAcontext precision ratio: 0.23 -> 1.54Improved
add entropy-based conflict resolution to the knowledge graph (ablation)TruthfulRAG (GPT-4o-mini) · SQuADaccuracy: 78.2 -> 81.1Improved
add entropy-based conflict resolution to the knowledge graph (ablation)TruthfulRAG (GPT-4o-mini) · FaithEvalaccuracy: 69.3 -> 69.5No clear effect
add entropy-based conflict resolution to the knowledge graph (ablation)TruthfulRAG (GPT-4o-mini) · RealtimeQAcontext precision ratio: 1.80 -> 1.54Mixed
replace natural-language context with KG structured reasoning pathsGPT-4o-mini · four datasetslog-probability of correct answerImproved
use TruthfulRAG instead of standard RAGGPT-4o-mini · MuSiQue-goldenaccuracy: 89.9 -> 93.2Improved
use KRE prompting on non-conflicting contextsGPT-4o-mini · SQuAD-goldenaccuracy: 97.9 -> 83.2Hurt
use TruthfulRAG instead of standard RAGGPT-4o-mini · RealtimeQAaccuracy: 67.3 -> 85.0Improved
use COIECD decoding instead of standard RAGGPT-4o-mini · FaithEvalaccuracy: 61.3 -> 53.9Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy