3 ms·
It looks like they go from tree -> sequence via prefix notation. I'm curious why Lample decided on this seq2seq approach when it seems that there might be model
by orange3xchicken 6y ago
It looks like they go from tree -> sequence via prefix notation. I'm curious why Lample decided on this seq2seq approach when it seems that there might be models which could be more naturally applied to the tree structure directly [1, 2]
[1] https://arxiv.org/abs/1902.07282 https://arxiv.org/abs/1902.07282 (an AST translation system)
[2] https://arxiv.org/abs/1609.02907 https://arxiv.org/abs/1609.02907 (GCNN)
- shmageggy 6y agoFor the same reason people use huge transformers for sentences in natural language (which are also tree structured): they scale really well. If you have enough data, huge transformers have huge capacity. If you notice, this paper is entirely about how to cleverly generate a massive dataset. There is no novelty in the model -- they just use a standard approach described in two paragraphs.