Pith. sign in

REVIEW 4 cited by

Exploring Chain-of-Thought Style Prompting for Text-to-SQL

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.14215 v2 pith:MIC2YDPY submitted 2023-05-23 cs.CL

Chang-You Tai , Ziru Chen , Tianshu Zhang , Xiang Deng , Huan Sun This is my paper · ORCID
classification cs.CL
keywords promptingtext-to-sqlparsingreasoningllmsmethodabilityabsolute
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In-context learning with large language models (LLMs) has recently caught increasing attention due to its superior few-shot performance on various tasks. However, its performance on text-to-SQL parsing still has much room for improvement. In this paper, we hypothesize that a crucial aspect of LLMs to improve for text-to-SQL parsing is their multi-step reasoning ability. Thus, we systematically study how to enhance LLMs' reasoning ability through chain of thought (CoT) style prompting, including the original chain-of-thought prompting (Wei et al., 2022b) and least-to-most prompting (Zhou et al., 2023). Our experiments demonstrate that iterative prompting as in Zhou et al. (2023) may be unnecessary for text-to-SQL parsing, and using detailed reasoning steps tends to have more error propagation issues. Based on these findings, we propose a new CoT-style prompting method for text-to-SQL parsing. It brings 5.2 and 6.5 point absolute gains on the Spider development set and the Spider Realistic set, respectively, compared to the standard prompting method without reasoning steps; 2.4 and 1.5 point absolute gains, compared to the least-to-most prompting method.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. SDE-SQL: Enhancing Text-to-SQL Generation in Large Language Models via Self-Driven Exploration with SQL Probes

    cs.CL 2025-06 conditional novelty 6.0 of 10

    SDE-SQL improves text-to-SQL accuracy by having the model generate and execute exploratory SQL probes to learn database contents before and while writing the final query.

  2. Toward Real-World Table Agents: Capabilities, Workflows, and Design Principles for LLM-based Table Intelligence

    cs.AI 2025-07 conditional novelty 4.0 of 10

    A capability-centric survey of LLM-based table agents, plus an experimental study showing that most Text-to-SQL agent methods fail to improve weaker open-source models on Spider and BIRD.

  3. Talking to Data: Designing Smart Assistants for Humanities Databases

    cs.CL 2025-06 conditional novelty 4.0 of 10

    A RAG-based smart assistant for the Prozhito diary archive combines hybrid retrieval and SQL filtering; DeepSeek-V3 scores highest on answer accuracy, but all tested models can be jailbroken by framing harmful questio...

  4. Meta-aware Learning in text-to-SQL Large Language Model

    cs.AI 2025-05 conditional novelty 4.0 of 10

    Combining schema, chain-of-thought, metadata knowledge, and tokenized prompt structures during fine-tuning improves text-to-SQL execution accuracy on private business databases compared to schema-only fine-tuning.

Pith tools