There's some papers on applying transformers to trees: https://arxiv.org/abs/1909.06639 , https://arxiv.org/abs/1911.09983 , https://papers.nips.cc/paper/2019/hash/6e0917469214d8fbd8c517dcdc6b8dcf-Abstract.html

And some recent work on tree extraction: https://arxiv.org/abs/2301.00447

There's also this paper which recovers a tree by observing the leaf nodes: https://arxiv.org/abs/2208.14924