English(EN) Robust Active Learning for Few-Shot Example Selection in Text-to-SQL

新的主动学习策略改进了文本到SQL的示例选择

作者 PulseAugur 编辑部 · [2 个来源] · 2026-06-08 19:56

研究人员开发了一种新的主动学习策略，用于在文本到SQL系统中选择少样本示例。该方法解决了诸如注释可靠性变化和查询嵌入中语义多样性需求等挑战。提出的分层贪婪算法优化了异方差互信息目标，提供了理论保证和经验证据，表明在保持准确性的同时减少了标注工作量。 AI

影响降低了开发专用文本到SQL系统的成本，可能加速其采用。

排序理由学术论文，详细介绍了在特定AI应用中主动学习的新方法。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv stat.ML TIER_1 English(EN) · Arash Pourhabib · 2026-06-10 04:00

Robust Active Learning for Few-Shot Example Selection in Text-to-SQL

arXiv:2606.10125v1 Announce Type: new Abstract: Few-shot example retrieval is the dominant paradigm for grounding large language models (LLMs) in domain-specific text-to-SQL systems. However, the quality of the annotated example bank directly governs system accuracy, and expert a…
arXiv stat.ML TIER_1 English(EN) · Arash Pourhabib · 2026-06-08 19:56

面向文本到SQL少样本示例选择的鲁棒主动学习

Few-shot example retrieval is the dominant paradigm for grounding large language models (LLMs) in domain-specific text-to-SQL systems. However, the quality of the annotated example bank directly governs system accuracy, and expert annotation is prohibitively expensive. We formali…