Language-model training and adaptation · MLD-1014
Rejection-sampling fine-tuning
Generating candidates, retaining those passing a quality criterion, and fine-tuning on the retained outputs.
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 on verified successful solutions.
Original caveat
RFT can also mean reinforcement fine-tuning in other contexts.
Context and aliases
Level: Intermediate. Stages: Training · Adaptation · Evaluation.
Tags: DL · SLM · LLM · NLP.
Aliases: RFT (rejection sampling).
Related concepts
Original references
- Transformers glossary ↗ · Consulted
- PEFT: LoRA conceptual guide ↗ · Consulted
- TRL: supervised fine-tuning ↗ · Reference link
- TRL documentation ↗ · Consulted
- Original reference URL ↗