arXiv cs.CL· Gaetano Rossiello, Alfio Gliozzo, Nandana Mihindukulasooriya, Erick Lee, Nabila Siregar, Lisa Amini·· 11 小时前AI 评分31
面向流式 Text-to-SQL 的自精炼提示词归纳方法
Self-Refining Prompt Induction for Streaming Text-to-SQL
AI 导读
研究者提出 Self-Refining Prompt Induction（SRPI），一种在模型权重固定前提下把生成失败转化为可复用方言指令的离线方法，由 Refiner 借助三个 LLM 评判器的参考反馈改写 Generator 提示词。
正文 · 原文
Abstract:Text-to-SQL assistants for streaming engines must handle dialect-specific types, windowing rules, and temporal semantics, including constructs poorly handled by generic prompts. We present Self-Refining Prompt Induction (SRPI), an offline method that converts generation failures into reusable dialect instructions while keeping model weights fixed. A Refiner revises the Generator's prompt using reference-conditioned feedback from three LLM judges; a sampled regression guard checks earlier successes, and a separate Compressor reduces prompt length. Inference uses the frozen prompt without retrieval or query repair. We evaluate cross-domain and cross-model transfer on Confluent Cloud Flink SQL using a controlled synthetic workload seeded by 30 expert-written examples, with 335 induction examples and 257 questions from held-out domains. The same induced prompt improves mean judge scores across seven Generators by 7.1-19.2 percentage points, including an increase from 85.3% to 97.5% for the induction model. A separate compression study reduces prompt length by 52.7% with a 0.19-percentage-point score decrease. An assistant using the induced prompt is in early access, with initial user feedback to inform subsequent refinement.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Databases (cs.DB) |
| Cite as: | arXiv:2611.00960 [cs.CL] |
| (or arXiv:2611.00960v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.00960 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Gaetano Rossiello [view email]
[v1]
Sat, 31 Oct 2026 14:02:40 UTC (345 KB)
来源：arXiv cs.CL · arxiv.org