Research map / Large language models / Agents, code and math
MCTS-SQL: Light-Weight LLMs Can Master the Text-to-SQL Through Monte Carlo Tree Search
Shuozhi Yuan, Liming Chen, Miaomiao Yuan, Zhao Jin
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 MCTS-Refiner (ablation) | MCTS-SQL · BIRD dev | EX: 16.8 -> 40.7 | Improved |
| add Selector (schema pruning) (ablation) | MCTS-SQL · BIRD dev | EX: 36.2 -> 40.7 | Improved |
| add Direct Generator (ablation) | MCTS-SQL · BIRD dev | EX: 34.8 -> 40.7 | Improved |
| add prefix-cache (ablation) | MCTS-SQL · BIRD dev | avg latency: 6.12s -> 2.84s | Improved |
| add prefix-cache (ablation) | MCTS-SQL · BIRD dev | EX: 40.69 -> 40.42 | No clear effect |
| add prefix-cache (ablation) | MCTS-SQL · BIRD dev | tokens: 2274 -> 864 | Improved |
| apply MCTS-SQL to Qwen-1.5B vs single-shot | Qwen2.5-Coder-1.5B-Instruct · BIRD dev | EX: 14.71 -> 40.69 | Improved |
| apply MCTS-SQL with a stronger base model | MCTS-SQL · BIRD dev (challenging) | EX: 22.78 -> 56.84 | Improved |
| use MCTS-SQL with GPT-4o instead of GPT-4o-mini | MCTS-SQL · Spider test | EX: 83.74 -> 86.63 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Alpha-SQL — Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search (ICML 2025) · better on 1 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 0.92)
- ROUTE: Robust Multitask Tuning and Collaboration for Text-to-SQL (ICLR 2025 · similarity 0.89)
- Structure-Guided Large Language Models for Text-to-SQL Generation (ICML 2025 · similarity 0.88)
- CogSQL: A Cognitive Framework for Enhancing Large Language Models in Text-to-SQL Translation (AAAI 2025 · similarity 0.88)
- Aligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMs (ICML 2026 · similarity 0.88)
- PARROT: A Benchmark for Evaluating LLMs in Cross-System SQL Translation (NeurIPS 2025 · similarity 0.87)
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.
