Tokenization and text representation · MLD-1185
Subword regularization
Training with alternative sampled subword segmentations to improve robustness.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Sample tokenizations during language-model training.
Original caveat
Training randomness need not be used at inference.
Context and aliases
Level: Intermediate. Stages: Data · Training · Inference.
Tags: NLP · SLM · LLM.
Aliases: None recorded.
Related concepts
Concepts linking here
Original references
- Hugging Face: chat templates ↗ · Reference link
- Transformers glossary ↗ · Consulted
- Hugging Face: tokenization algorithms ↗ · Consulted
- Original reference URL ↗