What Is Temperature? How It Affects Output Randomness
Scope note: This article explains the topic using public research and common engineering patterns. Retrieval, reranking, generation, and source-selec…
-
-
Top-K Sampling: Limiting the Next-Token Candidate Set
Top-K sampling is a generation control strategy: when predicting the next token, AI only selects from the K highest-probability candidates, ignoring e…
-
Does Temperature Affect Citations? Generation and Source Selection Are Different Stages
Scope note: This article explains the topic using public research and common engineering patterns. Retrieval, reranking, generation, and source-selec…
-
Autoregressive Generation: How AI Writes Its Answer One Word at a Time
Autoregressive generation is how AI produces text: not outputting a complete paragraph at once, but predicting one token at a time — generating a toke…
-
Multi-Head Attention: Representation Subspaces, Not Content Scores
Multi-head attention uses multiple sets of attention ("heads") simultaneously, each focusing on a different dimension — syntax, semantics, entity rela…
Have a GEO Question?
Can’t find what you need? Reach out — we’re happy to help.
English ▾
Chinese (Simplified)