AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

DCTR: Dual-Constraint Subgraph Optimization for Knowledge Graph-based Retrieval-Augmented Generation

Yukun Cao, Zirui Xu, Dongyang Li, Zhihao Guo, Luobin Huang, Lisheng Wang

AAAI 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 1 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 upper-bound path P_upper (ablation)DCTR · WebQSPMacro-F1: 70.33 -> 72.13Improved
add lower-bound path P_lower (ablation)DCTR · CWQMacro-F1: 46.92 -> 50.12Improved
add DSRS ranking loss (ablation)DCTR + Llama3.1-8B · WebQSPMacro-F1: 65.73 -> 72.13Improved
add DSRS ranking loss (ablation)DCTR retriever · CWQGPT-4o triple recall: 0.820 -> 0.839Improved
add P_lower (ablation)DCTR retriever · WebQSPpath triple recall: 0.851 -> 0.890Improved
replace SubgraphRAG retrieval with DCTR (8B)Llama3.1-8B · WebQSP-subMacro-F1: 71.92 -> 73.85Improved
replace SubgraphRAG retrieval with DCTR (70B)Llama3.1-70B · CWQ-subMacro-F1: 60.39 -> 63.27Improved
DCTR 8B vs GNN-RAG (fine-tuned)Llama3.1-8B · CWQMacro-F1: 59.4 -> 50.12Hurt
cross-dataset transfer (train on other dataset)DCTR + 8B · CWQMacro-F1: 50.12 -> 38.92Hurt
cross-dataset transfer (train on other dataset)DCTR + 8B · WebQSPMacro-F1: 72.13 -> 67.52Hurt

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