English(EN) When experts grade LLM answers in their own field, how well do the citations hold up? ExpertQA, a 2024 benchmark, has 484 experts write questions in their speci

ExpertQA基准测试揭示大型语言模型（LLM）引用不可靠

作者 PulseAugur 编辑部 · [1 个来源] · 2026-06-09 20:03

一个名为ExpertQA的新基准测试，于2024年开发，通过让484位专家在其专业领域提出问题来评估大型语言模型（LLM）。这些专家随后仔细评分LLM生成的答案，评估每个声明的支撑和可靠性。该基准测试显示，即使是写得很好的答案也常常包含未经证实的声明，而在医学领域，专家认为大约一半的引用来源不可靠。 AI

影响凸显了大型语言模型（LLM）在事实准确性和引用可靠性方面存在的重大问题，影响了在关键领域的信任和部署。

排序理由该集群描述了一个用于评估大型语言模型（LLM）的新学术基准测试。

在 Mastodon — fosstodon.org 阅读 →

ExpertQA
LLM

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

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一个用于评估大型语言模型（LLM）的新学术基准测试。
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, safety; 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: 113 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-06-09 20:03

当专家在自己的领域对大型语言模型（LLM）的回答进行评分时，引文的可靠性如何？2024年的ExpertQA基准测试，有484位专家在其专业领域撰写问题

When experts grade LLM answers in their own field, how well do the citations hold up? ExpertQA, a 2024 benchmark, has 484 experts write questions in their specialty, then judge the answers claim by claim. Even fluent answers leave many claims unsupported, and in medicine about ha…

链接 benjaminhan.net/…/20260609-expertqa

本文如何被排名

报道来源 [1]

当专家在自己的领域对大型语言模型（LLM）的回答进行评分时，引文的可靠性如何？2024年的ExpertQA基准测试，有484位专家在其专业领域撰写问题

相关实体

相关话题