Interested in memory saving for DL training?

Hi, I am the author of [Dynamic Tensor Rematerialization](https://arxiv.org/abs/2006.09616), a work that save memory for deep learning training, especially for dynamic shape/compute graph. I think this is a good fit for BladeDISC, and would you guys be interested in hearing a talk about the work?

No source files, tests, or implementation entry points are identified. Start by reviewing the Dynamic Tensor Rematerialization paper and the BladeDISC compiler architecture, then confirm whether the proposed work has a defined integration scope and acceptance criteria.

Written by the indexing model from the issue text.