83adc9225e4deb67d7ce42d58fe5157c-Supplemental.pdf
–Neural Information Processing Systems
Asforleaf nodes, their representations rli (l = 1) can be obtained by applying leaf transformation on the embeddings of their correspondingelementswti (e.g. A high-level actionG is indeed a sequence of M and C as (M1C1 MLCL), where L represents the highest layer. As for data splits, we split each dataset into the train set and the test set for all tasks according to previous works. More details about train and test sizes can be seen in Tab. 4.
Neural Information Processing Systems
Feb-9-2026, 04:53:30 GMT