C.W.K.
Stream
Quiz · 4 questions

🧩 Chunking and Ingestion

How you slice the corpus decides what you can find

Level 0Scout
0 XP0/41 lessons0/10 achievements
0/120 XP to next level120 XP to go0% complete

Quiz

01Why does embedding a whole 10-page document as one vector usually hurt retrieval quality?
02You split with a character-based splitter at 1200 chars. The same chunk is 240 tokens in English and 700 tokens in Korean. What is the consequence for an 8192-token model?
03When does structure-aware splitting (Markdown headers, code AST) outperform a generic recursive splitter?
04What is the single biggest production performance win that disciplined metadata enables?
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.