Research map / Large language models / Retrieval and knowledge
DAMR: Efficient and Adaptive Context-Aware Knowledge Graph Question Answering with LLM-Guided MCTS
Yingxu Wang, Shiqi Fan, Mengzhu Wang, Siyang Gao, Chao Wang, Nan Yin
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 the path evaluation module (ablation) | DAMR · WebQSP | Hits@1: 91.2 -> 94.0 | Improved |
| add the path evaluation module (ablation) | DAMR · CWQ | Hits@1: 74.3 -> 78.0 | Improved |
| add dynamic pseudo-path fine-tuning of the evaluator (ablation) | DAMR · WebQSP | Hits@1: 91.9 -> 94.0 | Improved |
| add dynamic pseudo-path fine-tuning of the evaluator (ablation) | DAMR · CWQ | F1: 73.0 -> 75.1 | Improved |
| use the trained Transformer scorer instead of GPT-4.1 as path evaluator (ablation) | DAMR · WebQSP | Hits@1: 92.5 -> 94.0 | Improved |
| use the trained Transformer scorer instead of GPT-4.1 as path evaluator (ablation) | DAMR · CWQ | F1: 72.4 -> 75.1 | Improved |
| use stronger planner LLM (Llama2-13B to GPT 4.1) (ablation) | DAMR · CWQ | F1: 69.5 -> 75.1 | Improved |
| invoke LLM only in expansion phase to select top-k relations | DAMR · WebQSP | tokens per question: 10,680 -> 3,931 (vs RwT) | Improved |
| increase top-k relations from 2 to 3 (ablation) | DAMR · CWQ | Hits@1: 76.6 -> 78.0 | Improved |
| increase top-k relations from 3 to 5 (ablation) | DAMR · WebQSP | Hits@1: 94.0 -> 93.9 | No clear effect |
| increase max path length L from 2 to 4 (ablation) | DAMR · CWQ | Hits@1: 77.4 -> 78.0 | Improved |
| increase max path length L beyond 3 (ablation) | DAMR · WebQSP | Hits@1: 94.0 -> 93.7 | No clear effect |
Closest papers by meaning
- Debate on Graph: A Flexible and Reliable Reasoning Framework for Large Language Models (AAAI 2025 · similarity 0.91)
- Harnessing Large Language Models for Knowledge Graph Question Answering via Adaptive Multi-Aspect Retrieval-Augmentation (AAAI 2025 · similarity 0.90)
- EPERM: An Evidence Path Enhanced Reasoning Model for Knowledge Graph Question and Answering (AAAI 2025 · similarity 0.90)
- Knowledge Editing with Dynamic Knowledge Graphs for Multi-Hop Question Answering (AAAI 2025 · similarity 0.90)
- Knowledge Graph Prompting for Multi-Document Question Answering (AAAI 2024 · similarity 0.90)
- Graph Reasoning Transformers for Knowledge-Aware Question Answering (AAAI 2024 · 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.
