Decoding and inference behavior · MLD-0779
Multi-token prediction
Predicting multiple future tokens through auxiliary heads or objectives rather than only one next token.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Train extra future-token prediction targets.
Original caveat
Training objectives and inference use vary across methods.
Context and aliases
Level: Intermediate. Stages: Inference · Evaluation · Deployment.
Tags: SLM · LLM · NLP.
Aliases: MTP.
Related concepts
Original references
- Hugging Face: generation strategies ↗ · Reference link
- Fast Inference from Transformers via Speculative Decoding ↗ · Consulted
- vLLM documentation ↗ · Consulted
- Original reference URL ↗