Brain-LLM Alignment Tracks Training Data, Not Typology
跨语言验证发现大脑语言网络与LLM的对齐主要受训练数据驱动,而非语言类型学差异。
arXiv:2605.23032v1 Announce Type: cross Abstract: Brain-LLM alignment is well established in English, yet the brain's language network is neuroanatomi…
跨语言验证发现大脑语言网络与LLM的对齐主要受训练数据驱动,而非语言类型学差异。
arXiv:2605.23032v1 Announce Type: cross Abstract: Brain-LLM alignment is well established in English, yet the brain's language network is neuroanatomi…
大模型与大脑在跨语言条件下的对齐鲁棒性,首次从计算根源上得到解释。
arXiv:2605.21049v1 Announce Type: new Abstract: Large language models (LLMs) reliably predict neural activity during language comprehension and transf…
被ICML 2026收录,提出针对低资源语言的LLM鲁棒多语言适配方法,增强跨语言泛化能力。
arXiv:2510.14466v3 Announce Type: replace Abstract: Large language models (LLMs) continue to struggle with low-resource languages, primarily due to li…
大模型面对同一概念的不同语言或编码形式时,会表现出“分隔化”现象,揭示其泛化能力的缺陷。
arXiv:2605.19284v1 Announce Type: cross Abstract: In the training data used by large language models (LLMs), the same latent concept is often presente…
临床AI系统在细微扰动和多语言场景下存在诊断崩溃风险,这篇系统性审计揭开了安全漏洞。
arXiv:2605.16993v1 Announce Type: cross Abstract: Current clinical artificial intelligence (AI) systems are evaluated almost exclusively on clean, sta…
揭示大模型英语偏见真相,证明持续预训练成本优势不存在,语言专用投资或成必然。
arXiv:2605.15613v1 Announce Type: new Abstract: Through an analysis of sequences generated by open-weight large language models (LLMs), we demonstrate…
请提供您想推荐的在线工具的名称或链接,我将为您生成推荐语、分类、标签和评分。
IT之家 5 月 18 日消息,据看看新闻 5 月 16 日报道,有网友爆料: 在建的京东上海总部大楼“歪”了 。记者在现场看到,建筑顶部脚手架的确发生倾斜。工程项目经理表示,建造平台爬升架的爬爪松脱,造成部分脚手架外架扭曲变形,主体结构未受影响,未造成人员伤亡。 5 月 17 日,普陀区建设工程安…