English(EN) Evaluating LLMs on Conversational Text-to-SQL under Chain Ambiguity and Intent Drift

新的TIDE-Bench基准评估了大型语言模型在对话式文本到SQL意图漂移方面的表现

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 04:00

研究人员推出了TIDE-Bench，一个旨在评估大型语言模型（LLMs）在对话式文本到SQL任务上的新基准，特别关注链歧义和意图漂移。该基准建立在BIRD数据集的514个锚点SQL之上，包含1,542个样本，并引入了用于识别链歧义和解决意图漂移的指标，超越了简单的执行准确性。使用TIDE-Bench对12个LLMs的评估揭示了在链识别方面存在重大挑战，以及在识别和解决用户意图漂移方面存在显著差距。 AI

影响该基准有望带来更强大的对话式AI系统，能够处理复杂的用户查询和修改。

排序理由该集群包含一篇介绍用于评估LLMs的新基准的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

新的TIDE-Bench基准评估了大型语言模型在对话式文本到SQL意图漂移方面的表现

本文如何被排名

Signal score: 22 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇介绍用于评估LLMs的新基准的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Yujia Liu, Jiayan Lin, Zijin Hong, Zheng Yuan, Shengyuan Chen, Hao Chen, Qinggang Zhang, Xiao Huang, Feiran Huang · 2026-09-01 04:00

评估LLMs在具有链歧义和意图漂移的对话文本到SQL任务上的表现

arXiv:2608.29543v1 Announce Type: cross Abstract: Recent advances in large language models (LLMs) have established conversational text-to-SQL as a practical interface between users and databases, often involving multiple turns of clarification and revision. However, existing benc…

本文如何被排名

报道来源 [1]

评估LLMs在具有链歧义和意图漂移的对话文本到SQL任务上的表现

相关实体

相关话题