Skip to main content
. Author manuscript; available in PMC: 2023 Aug 2.
Published in final edited form as: Adv Neural Inf Process Syst. 2022 Dec;35(DB):29776–29788.

Figure 5: Compositionality:

Figure 5:

We evaluate models’ capacity to reuse knowledge. (a) Models trained with a curriculum are compared to models trained from scratch. Models trained with a curriculum are overall more sample efficient. (b) Models trained on compositions are evaluated zero-shot on the respective elementary rules. Models fail overall to generalize from compositions to elementary rules.