mlsanityData to 14 Oct 2026

SPOC-SQL: Stage-wise Preference Optimization for Controllable Text-to-SQL

arXiv 2608.22772 ↗First posted 24 Aug 2026cs.CLcited by 0

On the map: LLM alignment / RLHF / preference optimisation → Process reward models for reasoning

Abstract (arXiv)

Text-to-SQL aims to translate natural language questions into executable SQL queries over relational databases, requiring multi-stage structured reasoning over database schemas and query constraints. However, existing methods treat this task as single-step generation, where models optimize entire SQL sequences without targeted feedback at key decision points and lack support for interacting with and controlling the intermediate generation process. To address this issue, we propose SPOC-SQL, which decomposes Text-to-SQL into four sequential subtasks following standard SQL execution logic and designs stage-specific optimization strategies for the model to learn key decisions. Specifically, we propose the implementation of fine-grained preference optimisation at key decision points across SQL stages, with the objective of enhancing structured decision-making during query construction. Furthermore, a structured decomposition strategy is designed, facilitating stage-wise intervention and correction through explicit intermediate representations. This results in more controllable and reliable SQL generation. Experiments demonstrate that incorporating stage-wise human knowledge consistently improves performance, validating the effectiveness of stage perception controllable generation.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Process reward models for reasoning, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 15 of its 22 references that are in the map. First posted on arXiv 24 Aug 2026.

Builds on

15 of 22 references.

References · 15

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  5. 2018Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Taskcited by 1,139
  6. 2023DIN-SQL: Decomposed In-Context Learning of Text-to-SQL with Self-Correctioncited by 477
  7. 2019RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parserscited by 424
  8. 2023Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluationcited by 385
  9. 2023RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQLcited by 195
  10. 2023C3: Zero-shot Text-to-SQL with ChatGPTcited by 164
  11. 2023MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQLcited by 164
  12. 2020Bridging Textual and Tabular Data for Cross-Domain Text-to-SQL Semantic Parsingcited by 133
3 more
  1. 2020Structure-Grounded Pretraining for Text-to-SQLcited by 126
  2. 2024MCS-SQL: Leveraging Multiple Prompts and Multiple-Choice Selection For Text-to-SQL Generationcited by 105
  3. 2023ACT-SQL: In-Context Learning for Text-to-SQL with Automatically-Generated Chain-of-Thoughtcited by 65

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025STaR-SQL: Self-Taught Reasoner for Text-to-SQL
  2. 2026ACE-SQL: Adaptive Co-Optimization via Empirical Credit Assignment for Text-to-SQL
  3. 2025Reinforcing Code Generation: Improving Text-to-SQL with Execution-Based Learning
  4. 2025JudgeSQL: Reasoning over SQL Candidates with Weighted Consensus Tournament
  5. 2025ExCoT: Optimizing Reasoning for Text-to-SQL with Execution Feedback
  6. 2025Uncovering the Impact of Chain-of-Thought Reasoning for Direct Preference Optimization: Lessons from Text-to-SQL
  7. 2026Learning When to Reason for Text-to-SQL via SFT and DPO
  8. 2025Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL