metadata
datasets:
- natolambert/skywork-preferences-80k-v0.1-cleaned
- allenai/preference-test-sets
MetaMetrics-RM-v1.0
- Authors Genta Indra Winata, David Anugraha, Lucky Susanto, Garry Kuwanto, Derry Tanti Wijaya
- Paper: https://arxiv.org/abs/2406.12845
- Model: meta-metrics/MetaMetrics-RM-v1.0
- Dataset:
- Code Repository: https://github.com/meta-metrics/metametrics
RewardBench LeaderBoard
| Model | Score | Chat | Chat Hard | Safety | Reasoning |
|---|---|---|---|---|---|
| nvidia/Llama-3.1-Nemotron-70B-Reward | 94.1 | 97.5 | 85.7 | 95.1 | 98.1 |
| meta-metrics/MetaMetrics-RM-v1.0 | 93.5 | 98.9 | 86.2 | 90.7 | 98.2 |
| SF-Foundation/TextEval-Llama3.1-70B | 93.5 | 94.1 | 90.1 | 93.2 | 96.4 |
| RLHFlow/ArmoRM-Llama3-8B-v0.1 | 90.4 | 96.9 | 76.8 | 90.5 | 97.3 |
Citation
If you find this work useful for your research, please consider citing:
@article{winata2024metametrics,
title={MetaMetrics: Calibrating Metrics For Generation Tasks Using Human Preferences},
author={Winata, Genta Indra and Anugraha, David and Susanto, Lucky and Kuwanto, Garry and Wijaya, Derry Tanti},
journal={arXiv preprint arXiv:2410.02381},
year={2024}
}