English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

TAHOE: Text-to-SQL with Automated Hint Optimization from Experience

Forum topic · 小凯 · 2026-06-12

Summary

TAHOE is a system that treats prompt optimization for Text-to-SQL as a dynamic data management problem, addressing the gap between LLM prototypes and production database deployments. It uses an error-driven hint learning pipeline spanning development and deployment phases, consolidating debugging traces into a structured Hint Bank. Compiler feedback is distilled into reusable Syntax Hints for dialect-specific rules, while execution and user feedback become Semantic Hints for schema- and user-specific logic. A policy layer models conflicting user intents as competing policies under shared natural-language triggers, with recency signals and post-learning attribution statistics. At inference, TAHOE retrieves relevant hints to guide LLM logical planning before SQL synthesis. On Spider 2.0-Snow with GPT-5.5 over 113 supervised examples, TAHOE raises pass rate from 61.95% to 79.42% and pass-at-4 from 72.57% to 87.61%, achieving 100% Snowflake syntax pass rate and cutting average compiler feedback rounds per candidate from 2.79 to 0.12—without updating model parameters. The same Hint Bank transfers to weaker backbones, yielding a 19.7 percentage-point gain on Doubao-2.0-lite. Paper: arXiv 2606.12387.

Overview

Field: ML Authors: Zhiyi Chen, Jie Song, Peng Li Published: 2026-06-10 arXiv: 2606.12387

Summary

Large Language Models (LLMs) have democratized database access through Text-to-SQL, but moving from prototypes to production remains difficult. Real deployments must handle strict SQL dialects, massive schemas, and evolving user preferences, while supervised fine-tuning is costly and rigid and agentic test-time scaling is expensive.

TAHOE treats prompt optimization as a dynamic data management problem. It uses an error-driven hint learning pipeline across Development and Deployment to consolidate debugging traces into a structured Hint Bank:

  • Syntax Hints: distilled from compiler feedback as reusable rules for dialect-specific SQL syntax.
  • Semantic Hints: derived from execution and user feedback for schema- and user-specific logic.
  • Policy layer: models conflicting user intents as competing policies under shared natural-language triggers, with recency signals and post-learning attribution statistics summarizing experiential success, harm, inaction, and support.
  • At inference time, TAHOE retrieves relevant hints and guides the LLM through logical planning followed by SQL synthesis. The current implementation and evaluation cover the development-phase workflow; deployment-time human-feedback updates are left to future work.

    Results

    On Spider 2.0-Snow (113 supervised Spider 2.0-Snow-0212 examples, GPT-5.5), without updating model parameters:

  • Pass rate: 61.95% → 79.42%
  • Pass-at-4: 72.57% → 87.61%
  • 100% Snowflake syntax pass rate
  • Average compiler-feedback critique rounds per sampled candidate: 2.79 → 0.12
  • The same Hint Bank transfers to weaker backbones, including a 19.7 percentage-point pass-rate improvement on Doubao-2.0-lite.
  • Links

  • Paper: arXiv 2606.12387

Tags

#text-to-sql#llm#prompt-optimization#databases#arxiv#snowflake#paper

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177981123