@inproceedings{zhang-etal-2026-cores,
    title = "{CORES}: Code-Oriented Reasoning for Complex Text-to-{SQL} and Generalizable {T}able{QA}",
    author = "Zhang, Meng  and
      Jin, Ruochun  and
      Peng, Yuanxi  and
      Yang, Wenjing  and
      Wang, Haotian  and
      Sun, Liting  and
      Hu, Kun  and
      Yang, Silin  and
      Ke-di, Zhang",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.237/",
    pages = "4827--4847",
    ISBN = "979-8-89176-395-1",
    abstract = "Text-to-SQL aims to bridge the gap between human intent and relational databases. While LLMs have shown proficiency in generating simple SQL queries, they struggle with complex analytical tasks. Moreover, models fine-tuned on SQL generation often suffer from catastrophic forgetting, which lose the versatility of procedural reasoning and pertaining to generation constraints. Inspired by the usage of high-resource programming languages as LLM reasoning intermediaries, we propose CORES model, which leverages Python as a procedural reasoning pivot to enhance both complex SQL generation and tabular reasoning. It decomposes complex queries into Python reasoning traces before generating the final SQL, which bridges the gap between procedural reasoning and declarative expression. In order to internalize this reasoning capability, we fine-tune LLMs via GRPO with tailored process reward functions that mitigate the sparse feedback problem. We experimentally verify the effectiveness of CORES on six text-to-SQL benchmarks, where ours outperforms baselines by 6.44{\%} on average, while maintains good capability on three tableQA benchmarks."
}