# Knowledge Distillation with Structured Chain-of-Thought for Text-to-SQL

arXiv: https://arxiv.org/abs/2512.17053 · HTML: https://mlsanity.com/paper/2512.17053/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 11 of its 38 references that are in the map.
- First posted on arXiv 18 Dec 2025.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)
- Knowledge distillation / [Knowledge distillation for LLMs](https://mlsanity.com/distill/knowledge-distillation-for-llms/)

## Builds on (most cited first, 11 of 11)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [C3: Zero-shot Text-to-SQL with ChatGPT](https://mlsanity.com/paper/2307.07306/) (2023)
- [Learning by Distilling Context](https://mlsanity.com/paper/2209.15189/) (2022)
- [A comprehensive evaluation of ChatGPT's zero-shot Text-to-SQL capability](https://mlsanity.com/paper/2303.13547/) (2023)
- [Evaluating the Text-to-SQL Capabilities of Large Language Models](https://mlsanity.com/paper/2204.00498/) (2022)
- [Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL](https://mlsanity.com/paper/2505.20315/) (2025)
- [PET-SQL: A Prompt-Enhanced Two-Round Refinement of Text-to-SQL with Cross-consistency](https://mlsanity.com/paper/2403.09732/) (2024)
- [ExCoT: Optimizing Reasoning for Text-to-SQL with Execution Feedback](https://mlsanity.com/paper/2503.19988/) (2025)
- [Divide and Prompt: Chain of Thought Prompting for Text-to-SQL](https://mlsanity.com/paper/2304.11556/) (2023)
- [Mixed Distillation Helps Smaller Language Model Better Reasoning](https://mlsanity.com/paper/2312.10730/) (2023)
- [Uncovering the Impact of Chain-of-Thought Reasoning for Direct Preference Optimization: Lessons from Text-to-SQL](https://mlsanity.com/paper/2502.11656/) (2025)

## Built on by (most cited first, 1 of 1)

- [LYNX: Analytical Program Synthesis via Grounding and Closed-Loop Refinement](https://mlsanity.com/paper/2610.14777/) (2026)

## Abstract (arXiv)

Deploying accurate Text-to-SQL systems at the enterprise level faces a difficult trilemma involving cost, security and performance. Current solutions force enterprises to choose between expensive, proprietary Large Language Models (LLMs) and low-performing Small Language Models (SLMs). Efforts to improve SLMs often rely on distilling reasoning from large LLMs using unstructured Chain-of-Thought (CoT) traces, a process that remains inherently ambiguous. Instead, we hypothesize that a formal, structured reasoning representation provides a clearer, more reliable teaching signal, as the Text-to-SQL task requires explicit and precise logical steps. To evaluate this hypothesis, we propose Struct-SQL, a novel Knowledge Distillation (KD) framework that trains an SLM to emulate a powerful large LLM. Consequently, we adopt a query execution plan as a formal blueprint to derive this structured reasoning. Our SLM, distilled with structured CoT, achieves an absolute improvement of 8.1% over an unstructured CoT distillation baseline. A detailed error analysis reveals that a key factor in this gain is a marked reduction in syntactic errors. This demonstrates that teaching a model to reason using a structured logical blueprint is beneficial for reliable SQL generation in SLMs.