---
title: "LitE-SQL — paper by Piao et al. (2026) — summary and ideas"
description: "LitE-SQL offers a compact two-part pipeline instead. 31 references. Among them: 3."
authors: ["Shengmin Piao", "Jieun Lee", "Sanghyun Park"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: EACL 2026"
source: "https://aclanthology.org/2026.findings-eacl.186/"
doi: "10.18653/v1/2026.findings-eacl.186"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/piao-2026-lite-sql-lightweight-efficient-text/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/piao-2026-lite-sql-lightweight-efficient-text/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/5/) / LitE-SQL: A Lightweight and Efficient Text-to-SQL Framework with Vector-based Schema Linking and Execution-Guided Self-Correction

# LitE-SQL: A Lightweight and Efficient Text-to-SQL Framework with Vector-based Schema Linking and Execution-Guided Self-Correction

**Authors:** [Shengmin Piao](https://md.research.elamur.ai/free/en/authors/shengmin-piao/), Jieun Lee and [Sanghyun Park](https://md.research.elamur.ai/free/en/authors/sanghyun-park/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: EACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.findings-eacl.186/) · [PDF](https://aclanthology.org/2026.findings-eacl.186.pdf) · [DOI 10.18653/v1/2026.findings-eacl.186](https://doi.org/10.18653/v1/2026.findings-eacl.186) · [OpenAlex](https://openalex.org/W7140126946)

**Citations:** 3 citations (OpenAlex, 2026-10-14)

## Summary

Text-to-SQL systems turn natural-language questions into database queries. Most strong systems now depend on large proprietary LLMs, which makes them costly to run and risky when data must stay private. LitE-SQL offers a compact two-part pipeline instead.

Its Schema Retriever embeds each column's metadata (name, description, table, value description) in advance and stores the vectors in a vector database. At inference only the question is encoded, and the top-k columns are fetched by cosine similarity. The embedding model is trained with HN-SupCon, a supervised contrastive loss that ignores easy negatives. It learns only from columns that look related but are not actually needed.

The SQL Generator is a small open model trained in two stages: supervised fine-tuning, then execution-guided reinforcement fine-tuning. This lets it repair its own queries using database error feedback rather than sampling many candidates.

On BIRD and Spider 1.0, LitE-SQL matches or beats LLM-based pipelines with far fewer parameters. This makes on-premises, privacy-preserving Text-to-SQL practical.

## Key points

- Execution accuracy (EX) is 72.10% on BIRD and 88.45% on Spider 1.0, using roughly 2× to 30× fewer parameters than the LLM-based methods it is compared against.
- HN-SupCon keeps only negatives scoring at least the positive's similarity minus a 0.1 margin. Applied to a Qwen3 embedding model, it beats both the vanilla model and standard SupCon: it raises TPR and schema-linking recall (SLR) while lowering FPR. The best settings were 8 negatives per question and k = 25 retrieved columns.
- The retriever has only 0.6B parameters, yet LitE-SQL trails the 200B-parameter multi-agent CHESS by just 0.68% EX. It also beats CodeS by 4.76% EX thanks to higher recall.
- Self-correction reduces every one of the four error categories tracked: syntax errors, references to nonexistent columns, references to nonexistent tables, and miscellaneous errors. The first correction pass on BIRD gives the largest gain, and the 7B generator eliminates all syntax errors on BIRD.
- Limitations: retrieving a fixed k columns still lets in false positives. Self-correction also cannot catch queries that run without error but are semantically wrong, such as wrong-table joins.
- Code: https://github.com/shengminp/LitE-SQL

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Yonsei University](https://md.research.elamur.ai/free/en/universities/yonsei-university/) — University, Seoul, South Korea · 17 works · 501 citations

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🌐 Computer Networks and Communications](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-networks-and-communications/) · 64 works

🧭 Topics (2): [🧾 Semantic parsing & meaning representation](https://md.research.elamur.ai/free/en/topics/semantic-parsing/) · 185 works · [💻 Code & programming languages](https://md.research.elamur.ai/free/en/topics/code-and-programming/) · 129 works

## Cite

Cite as: Shengmin Piao, Jieun Lee and Sanghyun Park. 2026. LitE-SQL: A Lightweight and Efficient Text-to-SQL Framework with Vector-based Schema Linking and Execution-Guided Self-Correction. Findings of the Association for Computational Linguistics: EACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{piao2026lite,
  title = {{LitE-SQL: A Lightweight and Efficient Text-to-SQL Framework with Vector-based Schema Linking and Execution-Guided Self-Correction}},
  author = {Shengmin Piao, Jieun Lee and Sanghyun Park},
  booktitle = {Findings of the Association for Computational Linguistics: EACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-eacl.186},
  url = {https://doi.org/10.18653/v1/2026.findings-eacl.186}
}
```

- [← previous Let's Simplify Step by Step: Guiding LLM Towards Multilingual Unsupervised Proficiency-Controlled Sentence Simplification](https://md.research.elamur.ai/free/en/papers/zhang-2026-let-s-simplify-step-step/)
- [next → LLM-Powered Benchmark Factory: Reliable, Generic, and Efficient](https://md.research.elamur.ai/free/en/papers/yuan-2026-llm-powered-benchmark-factory-reliable/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/5/)

## References

31 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). arXiv preprint arXiv:2303.08774.
2. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, and 1 others. 2020. [Language models are few-shot learners](https://md.research.elamur.ai/free/en/papers/brown-2020-language-models-few-shot-learners/). Advances in neural information processing systems, 33:1877–1901.
3. Dawei Gao, Haibin Wang, Yaliang Li, Xiuyu Sun, Yichen Qian, Bolin Ding, and Jingren Zhou. 2024. [Text-to-sql empowered by large language models: A benchmark evaluation](https://md.research.elamur.ai/free/en/papers/gao-2023-text-sql-empowered-large-language/). Proc. VLDB Endow., 17(5):1132–1145.
4. Zijin Hong, Zheng Yuan, Qinggang Zhang, Hao Chen, Junnan Dong, Feiran Huang, and Xiao Huang. 2025. Next-generation database interfaces: A survey of llm-based text-to-sql. IEEE Transactions on Knowledge and Data Engineering. [find](https://scholar.google.com/scholar?q=Zijin+Hong%2C+Zheng+Yuan%2C+Qinggang+Zhang%2C+Hao+Chen%2C+Junnan+Dong%2C+Feiran+Huang%2C+and+Xiao+Huang%2E+2025%2E+Next%2Dgeneration+database+interfaces%3A+A+survey+of+llm%2Dbased+text%2Dto%2Dsql%2E+IEEE+Transactions+on+Knowledg)
5. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. [LoRA: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). In International Conference on Learning Representations.
6. Binyuan Hui, Jian Yang, Zeyu Cui, Jiaxi Yang, Dayiheng Liu, Lei Zhang, Tianyu Liu, Jiajun Zhang, Bowen Yu, Keming Lu, and 1 others. 2024. Qwen2. 5-coder technical report. arXiv preprint arXiv:2409.12186. — [Qwen2.5-Coder Technical Report](https://md.research.elamur.ai/free/en/papers/hui-2024-qwen2-5-coder-technical-report/)
7. Prannay Khosla, Piotr Teterwak, Chen Wang, Aaron Sarna, Yonglong Tian, Phillip Isola, Aaron Maschinot, Ce Liu, and Dilip Krishnan. 2020. Supervised contrastive learning. Advances in neural information processing systems, 33:18661–18673. [find](https://scholar.google.com/scholar?q=Prannay+Khosla%2C+Piotr+Teterwak%2C+Chen+Wang%2C+Aaron+Sarna%2C+Yonglong+Tian%2C+Phillip+Isola%2C+Aaron+Maschinot%2C+Ce+Liu%2C+and+Dilip+Krishnan%2E+2020%2E+Supervised+contrastive+learning%2E+Advances+in+neural+information)
8. Dongjun Lee, Choongwon Park, Jaehyuk Kim, and Heesoo Park. 2025. [MCS-SQL: Leveraging multiple prompts and multiple-choice selection for text-to-SQL generation](https://md.research.elamur.ai/free/en/papers/lee-2025-mcs-sql-leveraging-multiple-prompts/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 337–353, Abu Dhabi, UAE.
9. Boyan Li, Yuyu Luo, Chengliang Chai, Guoliang Li, and Nan Tang. 2024a. The dawn of natural language to sql: Are we fully ready? Proc. VLDB Endow., 17(11):3318–3331. mitigating hallucinations in text-to-SQL generation. In Findings of the Association for Computational Linguistics: ACL 2024, pages 5456… [find](https://scholar.google.com/scholar?q=Boyan+Li%2C+Yuyu+Luo%2C+Chengliang+Chai%2C+Guoliang+Li%2C+and+Nan+Tang%2E+2024a%2E+The+dawn+of+natural+language+to+sql%3A+Are+we+fully+ready%3F+Proc%2E+VLDB+Endow%2E%2C+17%2811%29%3A3318%E2%80%933331%2E+mitigating+hallucinations+in+text%2Dt)
10. Fei Li and Hosagrahar V Jagadish. 2014. Constructing an interactive natural language interface for relational databases. Proceedings of the VLDB Endowment, 8(1):73–84. [find](https://scholar.google.com/scholar?q=Fei+Li+and+Hosagrahar+V+Jagadish%2E+2014%2E+Constructing+an+interactive+natural+language+interface+for+relational+databases%2E+Proceedings+of+the+VLDB+Endowment%2C+8%281%29%3A73%E2%80%9384%2E)
11. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. 2023. [Direct preference optimization: Your language model is secretly a reward model](https://md.research.elamur.ai/free/en/papers/rafailov-2023-direct-preference-optimization-language-model/). Advances in Neural Information Processing Systems, 36:53728– 53741.
12. Haoyang Li, Shang Wu, Xiaokang Zhang, Xinmei Huang, Jing Zhang, Fuxin Jiang, Shuai Wang, Tieying Zhang, Jianjun Chen, Rui Shi, Hong Chen, and Cuiping Li. 2025. Omnisql: Synthesizing high-quality text-to-sql data at scale. Proc. VLDB Endow., 18(11):4695–4709. [find](https://scholar.google.com/scholar?q=Haoyang+Li%2C+Shang+Wu%2C+Xiaokang+Zhang%2C+Xinmei+Huang%2C+Jing+Zhang%2C+Fuxin+Jiang%2C+Shuai+Wang%2C+Tieying+Zhang%2C+Jianjun+Chen%2C+Rui+Shi%2C+Hong+Chen%2C+and+Cuiping+Li%2E+2025%2E+Omnisql%3A+Synthesizing+high%2Dquality+text%2D)
13. Haoyang Li, Jing Zhang, Hanbing Liu, Ju Fan, Xiaokang Zhang, Jun Zhu, Renjie Wei, Hongyan Pan, Cuiping Li, and Hong Chen. 2024b. Codes: Towards building open-source language models for text-to-sql. Proceedings of the ACM on Management of Data, 2(3):1–28. [find](https://scholar.google.com/scholar?q=Haoyang+Li%2C+Jing+Zhang%2C+Hanbing+Liu%2C+Ju+Fan%2C+Xiaokang+Zhang%2C+Jun+Zhu%2C+Renjie+Wei%2C+Hongyan+Pan%2C+Cuiping+Li%2C+and+Hong+Chen%2E+2024b%2E+Codes%3A+Towards+building+open%2Dsource+language+models+for+text%2Dto%2Dsql%2E+Pr)
14. Jinyang Li, Binyuan Hui, Ge Qu, Jiaxi Yang, Binhua Li, Bowen Li, Bailin Wang, Bowen Qin, Ruiying Geng, Nan Huo, and 1 others. 2023. [Can llm already serve as a database interface? a big bench for large-scale database grounded text-to-sqls](https://md.research.elamur.ai/free/en/papers/li-2023-llm-already-serve-database-interface/). Advances in Neural Information Processing Systems, 36:42330–4…
15. Karime Maamari, Fadhil Abubaker, Daniel Jaroslawicz, and Amine Mhedhbi. 2024. The death of schema linking? text-to-SQL in the age of well-reasoned language models. In NeurIPS 2024 Third Table Representation Learning Workshop. [find](https://scholar.google.com/scholar?q=Karime+Maamari%2C+Fadhil+Abubaker%2C+Daniel+Jaroslawicz%2C+and+Amine+Mhedhbi%2E+2024%2E+The+death+of+schema+linking%3F+text%2Dto%2DSQL+in+the+age+of+well%2Dreasoned+language+models%2E+In+NeurIPS+2024+Third+Table+Represen)
16. Mohammadreza Pourreza, Hailong Li, Ruoxi Sun, Yeounoh Chung, Shayan Talaei, Gaurav Tarlok Kakkar, Yu Gan, Amin Saberi, Fatma Ozcan, and Sercan O Arik. 2025a. CHASE-SQL: Multi-path reasoning and preference optimized candidate selection in text-to-SQL. In The Thirteenth International Conference on Lea… [find](https://scholar.google.com/scholar?q=Mohammadreza+Pourreza%2C+Hailong+Li%2C+Ruoxi+Sun%2C+Yeounoh+Chung%2C+Shayan+Talaei%2C+Gaurav+Tarlok+Kakkar%2C+Yu+Gan%2C+Amin+Saberi%2C+Fatma+Ozcan%2C+and+Sercan+O+Arik%2E+2025a%2E+CHASE%2DSQL%3A+Multi%2Dpath+reasoning+and+prefer)
17. Mohammadreza Pourreza and Davood Rafiei. 2023. [Din-sql: Decomposed in-context learning of text-to-sql with self-correction](https://md.research.elamur.ai/free/en/papers/pourreza-2023-din-sql-decomposed-context-learning/). Advances in Neural Information Processing Systems, 36:36339–36348.
18. Hassan Shittu. 2024. Google gemini pro 1.5: All you need to know about this near perfect ai model. Fello AI Post. [find](https://scholar.google.com/scholar?q=Hassan+Shittu%2E+2024%2E+Google+gemini+pro+1%2E5%3A+All+you+need+to+know+about+this+near+perfect+ai+model%2E+Fello+AI+Post%2E)
19. Aditi Singh, Akash Shetty, Abul Ehtesham, Saket Kumar, and Tala Talaei Khoei. 2025. A survey of large language model-based generative ai for text-to-sql: Benchmarks, applications, use cases, and challenges. In 2025 IEEE 15th Annual Computing and Communication Workshop and Conference (CCWC), pages 00… [find](https://scholar.google.com/scholar?q=Aditi+Singh%2C+Akash+Shetty%2C+Abul+Ehtesham%2C+Saket+Kumar%2C+and+Tala+Talaei+Khoei%2E+2025%2E+A+survey+of+large+language+model%2Dbased+generative+ai+for+text%2Dto%2Dsql%3A+Benchmarks%2C+applications%2C+use+cases%2C+and+chall)
20. Shayan Talaei, Mohammadreza Pourreza, Yu-Chen Chang, Azalia Mirhoseini, and Amin Saberi. 2024. Chess: Contextual harnessing for efficient sql synthesis. CoRR, abs/2405.16755. — [arXiv:2405.16755](https://arxiv.org/abs/2405.16755) [find](https://scholar.google.com/scholar?q=Shayan+Talaei%2C+Mohammadreza+Pourreza%2C+Yu%2DChen+Chang%2C+Azalia+Mirhoseini%2C+and+Amin+Saberi%2E+2024%2E+Chess%3A+Contextual+harnessing+for+efficient+sql+synthesis%2E+CoRR%2C+abs%2F2405%2E16755%2E+%E2%80%94+%5BarXiv%3A2405%2E16755%5D%28http)
21. Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, and 1 others. 2023. [Gemini: a family of highly capable multimodal models](https://md.research.elamur.ai/free/en/papers/gemini-team-2023-gemini-family-highly-capable-multimodal/). arXiv preprint arXiv:2312.11805.
22. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. [Attention is all you need](https://md.research.elamur.ai/free/en/papers/vaswani-2017-attention-all-need/). Advances in neural information processing systems, 30.
23. Bing Wang, Changyu Ren, Jian Yang, Xinnian Liang, Jiaqi Bai, LinZheng Chai, Zhao Yan, Qian-Wen Zhang, Di Yin, Xing Sun, and Zhoujun Li. 2025. [MAC-SQL: A multi-agent collaborative framework for text-to-SQL](https://md.research.elamur.ai/free/en/papers/wang-2025-mac-sql-multi-agent-collaborative/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 540– 5…
24. Mohammadreza Pourreza and Davood Rafiei. 2024. DTS-SQL: Decomposed text-to-SQL with small large language models. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 8212–8220, Miami, Florida, USA. [find](https://scholar.google.com/scholar?q=Mohammadreza+Pourreza+and+Davood+Rafiei%2E+2024%2E+DTS%2DSQL%3A+Decomposed+text%2Dto%2DSQL+with+small+large+language+models%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+EMNLP+2024%2C+pages+8212%E2%80%9382)
25. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2023. [Self-consistency improves chain of thought reasoning in language models](https://md.research.elamur.ai/free/en/papers/wang-2022-self-consistency-improves-chain-thought/). In The Eleventh International Conference on Learning Representations.
26. Mohammadreza Pourreza, Shayan Talaei, Ruoxi Sun, Xingchen Wan, Hailong Li, Azalia Mirhoseini, Amin Saberi, and Sercan O Arik. 2025b. Reasoning-SQL: Reinforcement learning with SQL tailored partial rewards for reasoning-enhanced text-to-SQL. In Second Conference on Language Modeling. [find](https://scholar.google.com/scholar?q=Mohammadreza+Pourreza%2C+Shayan+Talaei%2C+Ruoxi+Sun%2C+Xingchen+Wan%2C+Hailong+Li%2C+Azalia+Mirhoseini%2C+Amin+Saberi%2C+and+Sercan+O+Arik%2E+2025b%2E+Reasoning%2DSQL%3A+Reinforcement+learning+with+SQL+tailored+partial+rew)
27. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. [Chain-of-thought prompting elicits reasoning in large language models](https://md.research.elamur.ai/free/en/papers/wei-2022-chain-thought-prompting-elicits-reasoning/). Advances in neural information processing systems, 35:24824– 24837.
28. Ge Qu, Jinyang Li, Bowen Li, Bowen Qin, Nan Huo, Chenhao Ma, and Reynold Cheng. 2024. [Before generation, align it!](https://md.research.elamur.ai/free/en/papers/wolf-2020-transformers-state-art-natural-language/) a novel and effective strategy for T Wolf. 2019. Huggingface’s transformers: State-ofthe-art natural language processing. arXiv preprint arXiv:1910.03771.
29. Dataset BIRD Spider 1.0 Dataset BIRD Spider 1.0 [find](https://scholar.google.com/scholar?q=Dataset+BIRD+Spider+1%2E0+Dataset+BIRD+Spider+1%2E0)
30. Tao Yu, Rui Zhang, Kai Yang, Michihiro Yasunaga, Dongxu Wang, Zifan Li, James Ma, Irene Li, Qingning Yao, Shanelle Roman, Zilin Zhang, and Dragomir Radev. 2018. [Spider: A large-scale human-labeled dataset for complex and cross-domain semantic parsing and text-to-SQL task](https://md.research.elamur.ai/free/en/papers/yu-2018-spider-large-scale-human-labeled/). In Proceedings of the 2018…
31. Yanzhao Zhang, Mingxin Li, Dingkun Long, Xin Zhang, Huan Lin, Baosong Yang, Pengjun Xie, An Yang, Dayiheng Liu, Junyang Lin, and 1 others. 2025. [Qwen3 embedding: Advancing text embedding and reranking through foundation models](https://md.research.elamur.ai/free/en/papers/zhang-2025-qwen3-embedding-advancing-text/). arXiv preprint arXiv:2506.05176.

*Updated 31 October 2026*