· Xiaojing Yang · NLP and LLMs
Transformers for NLP
How Transformers combine self-attention, feed-forward layers, residuals, and positional information.
How Transformers combine self-attention, feed-forward layers, residuals, and positional information.
Statistics is not just a set of formulas. It is a way to reason about uncertainty, evidence, and trust in AI experiments.
How discrete language becomes vector space, and why sentence embeddings matter for retrieval and evaluation.
ANOVA 问的是:组间差异是否大于组内噪声。
注意力机制是一种学习到的上下文选择方式:每个 token 应该看哪里。
偏差和方差解释了为什么模型太简单或太灵活都会失败。