PulseAugur
中
实时 11:10:34
English(EN) Do Open Language Models Invent Scripture? BILAL-BENCH: Benchmarking Islamic-Source Fabrication in Bengali

新基准揭示开放语言模型广泛捏造伊斯兰经文

作者 PulseAugur 编辑部 · [1 个来源] ·

一项名为BILAL-BENCH的新基准已被开发出来，用于评估开源语言模型捏造伊斯兰经文的能力，特别是在孟加拉语中。该基准包含240个涵盖《古兰经》经文、圣训和伊斯兰教法的问题，并经过专家验证。在对七种不同开放权重模型的1680个响应进行测试时，其中六种模型对79.3%至95.3%的问题自信地捏造了伊斯兰来源，模型大小没有带来明显改善。圣训出处被确定为所有评估模型的一个特别薄弱环节。 AI

影响 凸显了一个关键的安全问题，即开放语言模型可能生成听起来权威但实际上是错误的宗教文本，影响信任并可能导致错误信息。

排序理由 该集群描述了一篇介绍AI模型行为评估基准的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

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

新基准揭示开放语言模型广泛捏造伊斯兰经文

本文如何被排名

Signal score
1 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群描述了一篇介绍AI模型行为评估基准的学术论文。[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, safety, model release
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
1 days old
Coverage has settled into its steady-state source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.CL TIER_1 English(EN) · Sabah Nushra, Md Kishor Morol ·

    Do Open Language Models Invent Scripture? BILAL-BENCH: Benchmarking Islamic-Source Fabrication in Bengali

    arXiv:2610.40996v1 Announce Type: new Abstract: When a language model invents a Qur'anic verse or hadith, it gives false authority to text that readers may treat as sacred. Existing Islamic hallucination benchmarks are largely Arabic and often closed-ended, while Bengali resource…

相关实体

相关话题