Zenaique
Topics
Practice
Browse
Reference
Challenges
Pricing
Search…
⌘K
Topics
/
Tokenization
/
Unigram Lm
Unigram Lm
Subtopic
4 questions
Questions tagged with Unigram Lm — part of Tokenization.
Format
Difficulty
Role
Experience
Companies
Sort
Newest
Quality
Difficulty ↑
Difficulty ↓
Questions
BPE grows the vocabulary by adding merges. Unigram does the opposite. Explain the difference.
Flashcard
Easy
·
Qual 4.0
Explain the Unigram LM tokenization algorithm and how its training differs fundamentally from BPE. What does stochastic tokenization enable?
Short Answer
Hard
·
Qual 4.0
Google
Which statement correctly describes the Unigram Language Model tokenizer's training procedure?
Multiple Choice
Medium
·
Qual 4.0
Google
Spot the error: 'BPE finds the globally optimal vocabulary by iteratively merging the most frequent pair.'
Spot the Error
Medium
·
Qual 4.0