Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
Generative Verifiers: Reward Modeling as Next-Token Prediction
2025-01-22
· ICLR 2025 Poster ·
anchor
Findings
IC-121
Gemini 1.0 Pro, when prompted as a zero-shot chain-of-thought judge with majority voting, underperforms fine-tuned smaller Gemma models as verifiers on GSM8K