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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add knowledge-graph construction and retrieval to entropy filtering (ablation) | TruthfulRAG (GPT-4o-mini) · FaithEval | accuracy: 64.8 -> 69.5 | Improved |
| add knowledge-graph construction and retrieval to entropy filtering (ablation) | TruthfulRAG (GPT-4o-mini) · SQuAD | accuracy: 78.8 -> 81.1 | Improved |
| add knowledge-graph construction and retrieval to entropy filtering (ablation) | TruthfulRAG (GPT-4o-mini) · RealtimeQA | context precision ratio: 0.23 -> 1.54 | Improved |
| add entropy-based conflict resolution to the knowledge graph (ablation) | TruthfulRAG (GPT-4o-mini) · SQuAD | accuracy: 78.2 -> 81.1 | Improved |
| add entropy-based conflict resolution to the knowledge graph (ablation) | TruthfulRAG (GPT-4o-mini) · FaithEval | accuracy: 69.3 -> 69.5 | No clear effect |
| add entropy-based conflict resolution to the knowledge graph (ablation) | TruthfulRAG (GPT-4o-mini) · RealtimeQA | context precision ratio: 1.80 -> 1.54 | Mixed |
| replace natural-language context with KG structured reasoning paths | GPT-4o-mini · four datasets | log-probability of correct answer | Improved |
| use TruthfulRAG instead of standard RAG | GPT-4o-mini · MuSiQue-golden | accuracy: 89.9 -> 93.2 | Improved |
| use KRE prompting on non-conflicting contexts | GPT-4o-mini · SQuAD-golden | accuracy: 97.9 -> 83.2 | Hurt |
| use TruthfulRAG instead of standard RAG | GPT-4o-mini · RealtimeQA | accuracy: 67.3 -> 85.0 | Improved |
| use COIECD decoding instead of standard RAG | GPT-4o-mini · FaithEval | accuracy: 61.3 -> 53.9 | Hurt |
Closest papers by meaning
- A Systematic Exploration of Knowledge Graph Alignment with Large Language Models in Retrieval Augmented Generation (AAAI 2025 · similarity 0.91)
- Simple is Effective: The Roles of Graphs and Large Language Models in Knowledge-Graph-Based Retrieval-Augmented Generation (ICLR 2025 · similarity 0.90)
- Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented Generation (ICLR 2025 · similarity 0.90)
- Can Knowledge-Graph-based Retrieval Augmented Generation Really Retrieve What You Need? (NeurIPS 2025 · similarity 0.90)
- WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia (NeurIPS 2024 · similarity 0.90)
- Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection (ICLR 2024 · similarity 0.89)
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.
