EasyDeepLearn

Name three practical failure modes caused by tokenization.

medium

Answer

  • (1) Number-splitting: '1234' can tokenize into ['12', '34'] → poor arithmetic.
  • Fix: digit-level tokenizers or explicit tools.
  • (2) Code / whitespace fragility: leading spaces change tokens (hello  vs  hello)(\mathrm{hello}\;\mathrm{vs}\;\mathrm{hello}), breaking few-shot examples.
  • (3) Non-English bloat: languages without English-heavy training data use 2-5x more tokens per character → higher cost and shorter effective context.
  • (4) Rare characters produce multiple bytes each, wasting context.
Check yourself — multiple choice
  • Tokenization has no failure modes
  • Number splitting, whitespace fragility, non-English token bloat, rare-char explosion
  • Only affects Chinese
  • Solved by higher temperature

Real-world tokenization pain: numbers, whitespace, non-English bloat, rare chars.

#tokenization

Practise LLMs & GenAI

214 interview questions in this topic.

Related questions