Research map / Large language models / Retrieval and knowledge
Plan-Answer-Refine-on-Graph: Structured Planning and Self-Refinement for Large Language Model Reasoning on Knowledge Graphs
Yuxin Shi, Han Fu, Zhuo Li, Chenghao Liu, Xiaoxue Ren, Jianling Sun
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 self-refinement to PARoG (GPT-3.5) (ablation) | PARoG · CWQ | Hits@1: 69.2 -> 73.3 | Improved |
| add self-refinement to PARoG (GPT-3.5) (ablation) | PARoG · GrailQA | Hits@1: 78.9 -> 82.7 | Improved |
| add self-refinement to PARoG (GPT-3.5) (ablation) | PARoG · WebQSP | Hits@1: 88.0 -> 89.3 | Improved |
| add self-refinement to PARoG (GPT-4) (ablation) | PARoG · WebQSP | Hits@1: 89.7 -> 91.2 | Improved |
| add self-refinement to PARoG (GPT-4) (ablation) | PARoG · GrailQA | Hits@1: 85.5 -> 87.2 | Improved |
| add self-refinement to PARoG (GPT-4) (ablation) | PARoG · CWQ | Hits@1: 77.2 -> 79.3 | Improved |
| use SPARQL-supervised 8B planner instead of GPT-3.5 planner (ablation) | PARoG · CWQ | Hits@1: 65.2 -> 73.3 | Improved |
| use SPARQL-supervised 8B planner instead of Deepseek-R1 planner (ablation) | PARoG · CWQ | Hits@1: 68.7 -> 73.3 | Improved |
| use SPARQL-supervised 8B planner instead of Deepseek-R1 planner (ablation) | PARoG · WebQSP | Hits@1: 88.5 -> 89.3 | Improved |
| use SPARQL-supervised 8B planner instead of GPT-3.5 planner (ablation) | PARoG · GrailQA | Hits@1: 76.9 -> 82.7 | Improved |
| replace PoG with PARoG on complex query types | PoG · Superlative queries | accuracy: +20.5% | Improved |
| replace PoG planning with SPARQL-guided plan-answer-refine | PoG · Truncation rate | truncation rate (%): 38.82 -> 19.48 | Improved |
Closest papers by meaning
- Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning (ICLR 2024 · similarity 0.92)
- Plan-on-Graph: Self-Correcting Adaptive Planning of Large Language Model on Knowledge Graphs (NeurIPS 2024 · similarity 0.92)
- Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph (ICLR 2024 · similarity 0.91)
- Debate on Graph: A Flexible and Reliable Reasoning Framework for Large Language Models (AAAI 2025 · similarity 0.90)
- Simple is Effective: The Roles of Graphs and Large Language Models in Knowledge-Graph-Based Retrieval-Augmented Generation (ICLR 2025 · similarity 0.90)
- Graph-constrained Reasoning: Faithful Reasoning on Knowledge Graphs with Large Language Models (ICML 2025 · similarity 0.90)
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.
