ComPO
community
AI & ML interests
None defined yet.
Recent Activity
Organization Card
This repo contains all the models for paper -
ComPO: Preference Alignment via Comparison Oracles
https://arxiv.org/abs/2505.05465
Please cite
@inproceedings{
Chen-2025-compo,
title={ComPO: Preference Alignment via Comparison Oracles},
author={Peter Chen and Xi Chen and Wotao Yin and Tianyi Lin},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
year={2025},
url={https://openreview.net/forum?id=0lNwIIHWhZ}
}
models
18
ComparisonPO/Mistral-7B-Instruct-A0.21B-ComPO
7B
•
Updated
•
6
ComparisonPO/Mistral-7B-Instruct-ComPO-3300pert-300iter-2
7B
•
Updated
•
1
ComparisonPO/Mistral-7B-Instruct-ComPO-3300pert-300iter
7B
•
Updated
•
3
ComparisonPO/Gemma-2-9b-it-SimPO-ComPO
9B
•
Updated
•
4
ComparisonPO/Mistral-Instruct-7B-SimPO-ComPO
7B
•
Updated
•
2
ComparisonPO/Llama-3-Instruct-8B-SimPO-ComPO
8B
•
Updated
•
1
ComparisonPO/Llama-3-Base-8B-DPO-ComPO
8B
•
Updated
•
1
ComparisonPO/Llama-3-Instruct-8B-DPO-ComPO
8B
•
Updated
•
1
ComparisonPO/Llama-3-Base-8B-DPO_clean
8B
•
Updated
•
2
ComparisonPO/Mistral-Base-7B-DPO
7B
•
Updated
•
1
•
2
datasets
0
None public yet