layoutxlm to onnx question

[githublsk]

I use layoutxlm to training my data for downstream, when I convert the model which I trained to onnx by huggingface code layoutlmv2-to-onnx, it occurs below problem, can you give me some advice? it seems concat two different types causing this problem, but I do not modify any code just using run xfun for token classification, which confused me so much, I hope you can help us