How to accomplish the multi-dimensional fancy indexing in PyTorch?

Question

I have a 3D tensor ([batch_size, seq_length, hidden_dim]) and a 2D list ([batch_size, seq_length]).

I want to using list to accomplish the s