Research map / Graph learning / Graph self-supervised learning
SADGA: Structure-Aware Dual Graph Aggregation Network for Text-to-SQL
Ruichu Cai, Jinjie Yuan, Boyan Xu, Zhifeng Hao
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 SADGA instead of RATSQL with GloVe | Spider Text-to-SQL · Spider dev | exact match accuracy: 62.7 -> 64.7 | Improved |
| add SADGA instead of RATSQL with BERT-large | Spider Text-to-SQL · Spider test | exact match accuracy: 65.6 -> 66.7 | Improved |
| add SADGA instead of RATSQL with GAP | Spider Text-to-SQL · Spider test | exact match accuracy: 69.7 -> 70.1 | Improved |
| add Local Graph Linking (ablation) | SADGA · Spider dev Extra Hard | accuracy: 38.6 -> 42.8 | Improved |
| add Local Graph Linking (ablation) | SADGA · Spider dev Easy | accuracy: 83.5 -> 82.3 | Hurt |
| add Structure-Aware Aggregation (ablation) | SADGA · Spider dev All | accuracy: 62.9 -> 64.7 | Improved |
| add bidirectional graph aggregation (ablation) | SADGA · Spider dev All | accuracy: 62.9 -> 64.7; 61.5 -> 64.7 | Improved |
| use structure-aware aggregation rather than dual-graph encoding alone for Q-S linking (ablation) | SADGA · Spider dev All | accuracy: 63.1 -> 64.7 | Improved |
| represent relations with relation nodes rather than edge types (ablation) | SADGA · Spider dev All | accuracy: 62.1 -> 64.7 | Improved |
| add global pooling (ablation) | SADGA · Spider dev All | accuracy: 63.5 -> 64.7 | Improved |
| use a learned aggregation gate instead of fixed 0.5 weighting (ablation) | SADGA · Spider dev All | accuracy: 61.2 -> 64.7 | Improved |
| add cross-graph relation features to aggregation (ablation) | SADGA · Spider dev All | accuracy: 62.7 -> 64.7 | Improved |
Closest papers by meaning
- TrustUQA: A Trustful Framework for Unified Structured Data Question Answering (AAAI 2025 · similarity 0.87)
- Reasoning of Large Language Models over Knowledge Graphs with Super-Relations (ICLR 2025 · similarity 0.86)
- A Systematic Exploration of Knowledge Graph Alignment with Large Language Models in Retrieval Augmented Generation (AAAI 2025 · similarity 0.86)
- Simple is Effective: The Roles of Graphs and Large Language Models in Knowledge-Graph-Based Retrieval-Augmented Generation (ICLR 2025 · similarity 0.86)
- Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented Generation (ICLR 2025 · similarity 0.86)
- CogSQL: A Cognitive Framework for Enhancing Large Language Models in Text-to-SQL Translation (AAAI 2025 · similarity 0.86)
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.
