Yeah I agree with the intuition and hadn’t made the explicit connection to the shallow paths paper, thanks!
I would say that Edge Attribution Patching is the extreme form of this https://arxiv.org/abs/2310.10348 , where we just ignored almost all subgraphs H except for G \ {e_1} (removing one edge only), and still got reasonable results, and agrees with some more upcoming results.
Yeah I agree with the intuition and hadn’t made the explicit connection to the shallow paths paper, thanks!
I would say that Edge Attribution Patching is the extreme form of this https://arxiv.org/abs/2310.10348 , where we just ignored almost all subgraphs H except for G \ {e_1} (removing one edge only), and still got reasonable results, and agrees with some more upcoming results.