Why resize images in MMMU for DeepSeek-VL2?

[kydxh]

I notice that in evaluation of DeepSeek-VL2 on MMMU, the code resize the first image. But I don't know why. Could you please tell me the reasons?

(codes from "generate_inner" in VLMEvalKit/vlmeval/vlm/deepseek_vl2.py)

[kennymckormick]

The corresponding Pull Request was created by a member from the official DeepSeekVL2 team. We may need to seek help from the PR author @gnobitab

[kydxh]

Thank you very much.