| Paper | Code | chair_i | chair_s | ModelName | ReleaseDate |
|---|---|---|---|---|---|
| RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback | ✓ Link | 7.5 | 12.2 | RLHF-V | 2023-12-01 |
| RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V Trustworthiness | ✓ Link | 4.3 | 8.5 | RLAIF-V 7B | 2024-05-27 |
| RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V Trustworthiness | ✓ Link | 1.8 | 3.3 | RLAIF-V 12B | 2024-05-27 |