YanzhengXiang

Question about EXACT MATCH ACC

2

Hello, your work has made amazing progress on the spider data set, LGESQL + ELECTRA achieves 75.1 on the dev set, my question is whether this result is beam acc...

replicate experiments on the LLama model.

1

Hello, thank you for sharing your code. Now I want to replicate your experiments on the LLama model, mainly to experiment on the compound_words dataset. However, the data you provided...

Why using Using collapsed_tree?

2

I conducted experiments on NarrativeQA benchmark using the LLama-3 model, and the model used collapsed_tree for nearly all questions and I don't know the reason.