精选理由
微软研究搞了个新招，把黑盒模型怎么处理语言变成可以验证的假设，还真的用脑扫描去测，看哪些脑区在干活，挺有意思。
微软研究人员提出一种名为generative causal testing的方法，将黑盒语言模型转化为清晰假设，并通过fMRI脑部扫描进行验证。实验揭示了特定脑区对语言特征（如词义、句法）的响应模式，例如左侧颞叶对语义角色的敏感度。该方法在多个基准测试中优于传统解释性技术，为理解神经语言处理提供了新途径。
原文 · Microsoft Research
Researchers introduce generative causal testing, which translates black box models into clear hypoth...
Researchers introduce generative causal testing, which translates black box models into clear hypotheses and verifies them in the scanner, revealing what specific brain regions respond to in language. msft.it/6011vUtRd Your browser does not support the video tag. 🔗 View on Twitter 💬 0 🔄 1 ❤️ 6 👀 2273 📊 1 ⚡