wassname/SimPO
1
0
Fork 0
mirror of https://github.com/wassname/SimPO.git synced 2026-06-27 17:46:46 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Yu Meng
2024-07-09 14:55:28 -04:00
committed by GitHub
parent 9ef5dcbd69
commit 995fbaf260
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+2 -2
View File
@@ -7,8 +7,8 @@ This repository contains the code and released models for our paper [SimPO: Simp
## 🆕 Changelog
- [2024.07.08] We updated our paper ([v2](https://arxiv.org/abs/2405.14734v2))
- Additional baselines (RRHF, SLiC-HF, CPO)
- New Llama3-Instruct setting (v0.2) with [ArmoRM](https://huggingface.co/RLHFlow/ArmoRM-Llama3-8B-v0.1) as the preference label annotator, yielding better-performing models!
- [SimPO trainer](https://github.com/princeton-nlp/SimPO/blob/main/scripts/simpo_trainer.py) update for better reproducibility
- New Llama3-Instruct setting (v0.2) with [ArmoRM](https://huggingface.co/RLHFlow/ArmoRM-Llama3-8B-v0.1) as the preference label annotator, yielding better-performing models ([training script](https://github.com/princeton-nlp/SimPO/blob/main/training_configs/llama-3-8b-instruct-simpo-v2.yaml))!
- [SimPO trainer](https://github.com/princeton-nlp/SimPO/blob/main/scripts/simpo_trainer.py) update for better reproducibility. The hyperparameter `gamma` changed to `gamma_beta_ratio` for easier tuning.
## 🔗 Quick Links
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](#simple-preference-optimization-simpo)