[BUG] Qwen-VL-Max: incorrect text detection

[thdoan]

是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?

• [x] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions

该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?

• [x] 我已经搜索过FAQ | I have searched FAQ

当前行为 | Current Behavior

Is this the right place to report bugs about incorrect text detections?

Example:

期望行为 | Expected Behavior

9Y548

复现方法 | Steps To Reproduce

1. Go to https://huggingface.co/spaces/Qwen/Qwen-VL-Max
2. Upload this image:
3. Submit prompt "Reply with only the text detected in this image."

运行环境 | Environment

- OS: N/A
- Python: N/A
- Transformers: N/A
- PyTorch: N/A
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`): N/A

备注 | Anything else?

No response