Training
Reward model
A reward model scores outputs by quality.
Quick definition
A reward model scores outputs by quality.
- Category: Training
- Focus: model adaptation
- Used in: Adapting a base model to your domain or style.
What it means
It is used to optimize responses in training. In training workflows, reward model often shapes model adaptation.
How it works
Training adapts models through fine-tuning or preference optimization. It uses curated datasets and evaluation loops.
Why it matters
Training methods tailor models to your domain and use case.
Common use cases
- Adapting a base model to your domain or style.
- Improving instruction following for specific tasks.
- Reducing errors with better training data.
Example
Score helpfulness for RLHF.
Pitfalls and tips
Low-quality data can degrade performance. Keep datasets clean, representative, and well-labeled.
In BoltAI
In BoltAI, this is referenced when discussing model customization.
