Reward Modeling
Alignment TrainingactiveTraining neural networks on human preference comparisons to provide scalable reward signals for RL fine-tuning.
Key Papers
1
Tags
function:specificationstage:trainingscope:technique
Key Papers & Resources1
SEMINAL
Deep Reinforcement Learning from Human Preferences
Christiano et al.2017
Sub-Areas1
| Name | Status | Orgs | Papers |
|---|---|---|---|
| Process SupervisionStep-level reward signals for reasoning verification, as opposed to outcome-only rewards. | active | 0 | 1 |