# Compositional generalization through abstract representations in human and artificial neural networks

arXiv: https://arxiv.org/abs/2209.07431 · HTML: https://mlsanity.com/paper/2209.07431/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Sep 2022.

## Built on by (most cited first, 22 of 22)

- [Artificial General Intelligence (AGI)-Native Wireless Systems: A Journey Beyond 6G](https://mlsanity.com/paper/2405.02336/) (2024)
- [When does compositional structure yield compositional generalization? A kernel theory](https://mlsanity.com/paper/2405.16391/) (2024)
- [A Survey on Compositional Learning of AI Models: Theoretical and Experimental Practices](https://mlsanity.com/paper/2406.08787/) (2024)
- [Temporal Representation Alignment: Successor Features Enable Emergent Compositionality in Robot Instruction Following](https://mlsanity.com/paper/2502.05454/) (2025)
- [Continual task learning in natural and artificial agents](https://mlsanity.com/paper/2210.04520/) (2022)
- [Vector-based Representation is the Key: A Study on Disentanglement and Compositional Generalization](https://mlsanity.com/paper/2305.18063/) (2023)
- [Quantifying artificial intelligence through algorithmic generalization](https://mlsanity.com/paper/2411.05943/) (2024)
- [Banyan: Improved Representation Learning with Explicit Structure](https://mlsanity.com/paper/2407.17771/) (2024)
- [Algebras of actions in an agent's representations of the world](https://mlsanity.com/paper/2310.01536/) (2023)
- [Modularity is the Bedrock of Natural and Artificial Intelligence](https://mlsanity.com/paper/2602.18960/) (2026)
- [Uncertainty-guided Compositional Alignment with Part-to-Whole Semantic Representativeness in Hyperbolic Vision-Language Models](https://mlsanity.com/paper/2603.22042/) (2026)
- [NeuReasoner: Theory-grounded Mapping of Reasoning Elicitation Boundaries](https://mlsanity.com/paper/2606.29971/) (2026)
- [Bridging Neuroscience and AI: Environmental Enrichment as a Model for Forward Knowledge Transfer](https://mlsanity.com/paper/2405.07295/) (2024)
- [Inductive Biases for Zero-shot Systematic Generalization in Language-informed Reinforcement Learning](https://mlsanity.com/paper/2501.15270/) (2025)
- [Modularity Trumps Invariance for Compositional Robustness](https://mlsanity.com/paper/2306.09005/) (2023)
- [Development of Compositionality and Generalization through Interactive Learning of Language and Action of Robots](https://mlsanity.com/paper/2403.19995/) (2024)
- [Data Distributional Properties As Inductive Bias for Systematic Generalization](https://mlsanity.com/paper/2502.20499/) (2025)
- [Hybrid Learners Do Not Forget: A Brain-Inspired Neuro-Symbolic Approach to Continual Learning](https://mlsanity.com/paper/2503.12635/) (2025)
- [Feature, Alignment, and Supervision in Category Learning: A Comparative Approach with Children and Neural Networks](https://mlsanity.com/paper/2602.03124/) (2026)
- [BRo-JEPA: Learning Modular Transformations in Latent Space](https://mlsanity.com/paper/2606.01372/) (2026)
- [How are linear representations learned? Exact solutions to the dynamics of abstraction](https://mlsanity.com/paper/2607.08843/) (2026)
- [CoGe-GCD: Reframing Generalized Category Discovery with Compositional Generalization](https://mlsanity.com/paper/2609.10158/) (2026)

## Abstract (arXiv)

Humans have a remarkable ability to rapidly generalize to new tasks that is difficult to reproduce in artificial learning systems. Compositionality has been proposed as a key mechanism supporting generalization in humans, but evidence of its neural implementation and impact on behavior is still scarce. Here we study the computational properties associated with compositional generalization in both humans and artificial neural networks (ANNs) on a highly compositional task. First, we identified behavioral signatures of compositional generalization in humans, along with their neural correlates using whole-cortex functional magnetic resonance imaging (fMRI) data. Next, we designed pretraining paradigms aided by a procedure we term em primitives pretraining to endow compositional task elements into ANNs. We found that ANNs with this prior knowledge had greater correspondence with human behavior and neural compositional signatures. Importantly, primitives pretraining induced abstract internal representations, excellent zero-shot generalization, and sample-efficient learning. Moreover, it gave rise to a hierarchy of abstract representations that matched human fMRI data, where sensory rule abstractions emerged in early sensory areas, and motor rule abstractions emerged in later motor areas. Our findings give empirical support to the role of compositional generalization in human behavior, implicate abstract representations as its neural implementation, and illustrate that these representations can be embedded into ANNs by designing simple and efficient pretraining procedures.