Checklist

• 1. I have searched related issues but cannot get the expected help.
• 2. The bug has not been fixed in the latest version.
• 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.

Describe the bug

MPO训练时在trainer_dpo.py脚本中缺少对 self.is_vision_model、self.aux_loss_enabled等变量的定义，导致在 MultimodalDPOTrainer 类中无法使用这些属性。

Reproduction

File "/code/InternVL/internvl_chat/internvl/train/trainer_dpo.py", line 193, in get_batch_loss_metrics
forward_output = self.concatenated_forward(model, batch)
File "/code/InternVL/internvl_chat_mpo/internvl/train/trainer_dpo.py", line 100, in concatenated_forward
is_vision_model=self.is_vision_model,

Environment

transformers             4.45.0
trl                      0.9.2

Error traceback

No response