---
title: "SQLoom: Set-Contrastive Learning of Syntactic, Semantic, and Logical Dependencies for Text-to-SQL"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.03298v1"
permalink: "https://feed.astralor.com/items/itm_322d1e5cf63659d591730f86"
published: "2026-11-02T15:33:40.000Z"
score: 62
channel: "tech"
tags: ["LLM","AI Agent","OpenAI","安全","多模态","SQLoom","Set-Contrastive Learning","Syntactic"]
---

# SQLoom: Set-Contrastive Learning of Syntactic, Semantic, and Logical Dependencies for Text-to-SQL

## 速览

Text-to-SQL translates natural language questions into SQL queries, enabling database access without writing SQL. Fine-tuning open-weight large language models (LLMs) is cost-effective, but existing methods treat SQL as…

## 关键点

- Text-to-SQL translates natural language questions into SQL queries, enabling database access without writing SQL.
- Fine-tuning open-weight large language models (LLMs) is cost-effective, but existing methods treat SQL as ordinary text and optimize token-level likelihood without explicitly modeling its syntactic, semantic, and logica…
- Thus, a single incorrect alias or comparison operator can invalidate a query while contributing little to training loss.

## 全文

Text-to-SQL translates natural language questions into SQL queries, enabling database access without writing SQL. Fine-tuning open-weight large language models (LLMs) is cost-effective, but existing methods treat SQL as ordinary text and optimize token-level likelihood without explicitly modeling its syntactic, semantic, and logical requirements. Thus, a single incorrect alias or comparison operator can invalidate a query while contributing little to training loss. We propose SQLoom, a constraint-aware fine-tuning framework with supervision at three levels: syntactic validity under SQL grammar and database schema, semantic alignment between query decisions and the question, and logical consistency across dependent clause decisions. SQLoom extends supervised fine-tuning with a unified set-contrastive objective favoring sets of correct outputs over minimally different incorrect ones at each level. LLM agents propose this supervision, which a SQL parser and execution on counterexample databases verify. We also release Spider-Refined, a training dataset correcting 378 flawed Spider annotations and enriching database schemas. Experiments on Spider, BIRD, and three perturbed Spider variants show that SQLoom improves execution accuracy for Qwen3-1.7B from 62.9% zero-shot to 88.7% on Spider development, reaching 89.2% on Spider test and 73.4% on BIRD development. It outperforms fine-tuning, reinforcement learning, and agent-based baselines with up to 70 billion parameters. On question perturbations, SQLoom loses at most 0.6 accuracy points versus up to 12.1 for agent-based methods. It generates SQL in a single pass at 0.42s per query, 23$\times$ faster than a GPT-4-based agent method. These results demonstrate the effectiveness of set-contrastive supervision over token-level training.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_322d1e5cf63659d591730f86