Research map / Large language models / Language, safety and interpretability
Addressing Semantic Blind Spots in Text-to-SQL via Component Pre-generation and AST Matching Rewards
Xingyu Ma, Xin Tian, Lingxiang Wu, Xuepeng Wang, Zhilin Zhang, Xueming Tang, Jinqiao Wang
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 SQL component pre-generation reward (ablation) | base reward GRPO · BIRD-DEV (challenging) | EX: 55.56 -> 59.03 | Improved |
| add AST matching reward (ablation) | base reward GRPO · BIRD-DEV | EX (total): 63.89 -> 64.02 | Improved |
| add IDF node weighting to AST reward (ablation) | base + AST reward · BIRD-DEV | EX (total): 64.02 -> 64.21 | Improved |
| add IDF weighting to pre-gen + AST reward (ablation) | pre-gen + AST · BIRD-DEV (challenging) | EX: 60.42 -> 63.19 | Improved |
| add IDF weighting to pre-gen + AST reward (ablation) | pre-gen + AST · BIRD-DEV | EX (total): 64.99 -> 65.78 | Improved |
| replace n-gram with AST matching reward (ablation) | pre-generation · BIRD-DEV | EX (total): 63.80 -> 64.99 | Improved |
| add pre-generation with n-gram matching vs. base alone (ablation) | base reward GRPO · BIRD-DEV | EX (total): 63.89 -> 63.80 | No clear effect |
| apply pre-generation to all questions (remove complexity analysis) (ablation) | full method · BIRD-DEV | EX (total): 65.78 -> 64.86 | Hurt |
| use rule-based instead of model&rule complexity (ablation) | full method · BIRD-DEV | EX (total): 65.78 -> 65.25 | Hurt |
| use model-based complexity only (ablation) | full method · BIRD-DEV | EX (total): 65.78 -> 64.73 | Hurt |
| add n-gram local matching reward (ablation) | w/o matching reward · SPIDER-DEV | EM: 11.7 -> 37.2 | Improved |
| replace n-gram with AST matching reward (ablation) | local matching reward · SPIDER-TEST | EM: 36.3 -> 45.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- Alpha-SQL — Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search (ICML 2025) · better on 0 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search (ICML 2025 · similarity 0.89)
- Structure-Guided Large Language Models for Text-to-SQL Generation (ICML 2025 · similarity 0.89)
- ROUTE: Robust Multitask Tuning and Collaboration for Text-to-SQL (ICLR 2025 · similarity 0.87)
- SQLens: An End-to-End Framework for Error Detection and Correction in Text-to-SQL (NeurIPS 2025 · similarity 0.87)
- CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQL (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.
