# Improving Systematic Generalization Through Modularity and Augmentation

arXiv: https://arxiv.org/abs/2202.10745 · HTML: https://mlsanity.com/paper/2202.10745/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 51 references that are in the map.
- First posted on arXiv 22 Feb 2022.

## Builds on (most cited first, 27 of 27)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning](https://mlsanity.com/paper/1612.06890/) (2016)
- [A Survey of Data Augmentation Approaches for NLP](https://mlsanity.com/paper/2105.03075/) (2021)
- [Neural Module Networks](https://mlsanity.com/paper/1511.02799/) (2015)
- [Measuring Compositional Generalization: A Comprehensive Method on Realistic Data](https://mlsanity.com/paper/1912.09713/) (2019)
- [COGS: A Compositional Generalization Challenge Based on Semantic Interpretation](https://mlsanity.com/paper/2010.05465/) (2020)
- [Good-Enough Compositional Data Augmentation](https://mlsanity.com/paper/1904.09545/) (2019)
- [Task-Driven Modular Networks for Zero-Shot Compositional Learning](https://mlsanity.com/paper/1905.05908/) (2019)
- [A Benchmark for Systematic Generalization in Grounded Language Understanding](https://mlsanity.com/paper/2003.05161/) (2020)
- [Compositional generalization through meta sequence-to-sequence learning](https://mlsanity.com/paper/1906.05381/) (2019)
- [Compositional Generalization in Semantic Parsing: Pre-training vs. Specialized Architectures](https://mlsanity.com/paper/2007.08970/) (2020)
- [Improving Coherence and Consistency in Neural Sequence Models with Dual-System, Neuro-Symbolic Reasoning](https://mlsanity.com/paper/2107.02794/) (2021)
- [A causal view of compositional zero-shot recognition](https://mlsanity.com/paper/2006.14610/) (2020)
- [Learning Compositional Rules via Neural Program Synthesis](https://mlsanity.com/paper/2003.05562/) (2020)
- [Linguistic generalization and compositionality in modern artificial neural networks](https://mlsanity.com/paper/1904.00157/) (2019)
- [Compositional generalization in a deep seq2seq model by separating syntax and semantics](https://mlsanity.com/paper/1904.09708/) (2019)
- [Rearranging the Familiar: Testing Compositional Generalization in Recurrent Networks](https://mlsanity.com/paper/1807.07545/) (2018)
- [Independent Prototype Propagation for Zero-Shot Compositionality](https://mlsanity.com/paper/2106.00305/) (2021)
- [Jump to better conclusions: SCAN both left and right](https://mlsanity.com/paper/1809.04640/) (2018)
- [Inducing Transformer's Compositional Generalization Ability via Auxiliary Sequence Prediction Tasks](https://mlsanity.com/paper/2109.15256/) (2021)
- [Compositional Networks Enable Systematic Generalization for Grounded Language Understanding](https://mlsanity.com/paper/2008.02742/) (2020)
- [Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering](https://mlsanity.com/paper/2007.00266/) (2020)
- [Systematic Generalization on gSCAN with Language Conditioned Embedding](https://mlsanity.com/paper/2009.05552/) (2020)
- [How Modular Should Neural Module Networks Be for Systematic Generalization?](https://mlsanity.com/paper/2106.08170/) (2021)
- [Systematic Generalization on gSCAN: What is Nearly Solved and What is Next?](https://mlsanity.com/paper/2109.12243/) (2021)
- [Can Transformers Jump Around Right in Natural Language? Assessing Performance Transfer from SCAN](https://mlsanity.com/paper/2107.01366/) (2021)
- [Think before you act: A simple baseline for compositional generalization](https://mlsanity.com/paper/2009.13962/) (2020)

## Built on by (most cited first, 13 of 13)

- [Machine Psychology](https://mlsanity.com/paper/2303.13988/) (2023)
- [Position: Understanding LLMs Requires More Than Statistical Generalization](https://mlsanity.com/paper/2405.01964/) (2024)
- [Code-Switched Text Synthesis in Unseen Language Pairs](https://mlsanity.com/paper/2305.16724/) (2023)
- [Instilling Inductive Biases with Subnetworks](https://mlsanity.com/paper/2310.10899/) (2023)
- [Quantifying artificial intelligence through algorithmic generalization](https://mlsanity.com/paper/2411.05943/) (2024)
- [Compositional Generalization in Grounded Language Learning via Induced Model Sparsity](https://mlsanity.com/paper/2207.02518/) (2022)
- [Learning to Compose Representations of Different Encoder Layers towards Improving Compositional Generalization](https://mlsanity.com/paper/2305.12169/) (2023)
- [Improved Compositional Generalization by Generating Demonstrations for Meta-Learning](https://mlsanity.com/paper/2305.13092/) (2023)
- [Layer-wise Representation Fusion for Compositional Generalization](https://mlsanity.com/paper/2307.10799/) (2023)
- [D3: Data Diversity Design for Systematic Generalization in Visual Question Answering](https://mlsanity.com/paper/2309.08798/) (2023)
- [Make Haste Slowly: A Theory of Emergent Structured Mixed Selectivity in Feature Learning ReLU Networks](https://mlsanity.com/paper/2503.06181/) (2025)
- [A Compressive-Expressive Communication Framework for Compositional Representations](https://mlsanity.com/paper/2501.19182/) (2025)
- [OC-NMN: Object-centric Compositional Neural Module Network for Generative Visual Analogical Reasoning](https://mlsanity.com/paper/2310.18807/) (2023)

## Abstract (arXiv)

Systematic generalization is the ability to combine known parts into novel meaning; an important aspect of efficient human learning, but a weakness of neural network learning. In this work, we investigate how two well-known modeling principles -- modularity and data augmentation -- affect systematic generalization of neural networks in grounded language learning. We analyze how large the vocabulary needs to be to achieve systematic generalization and how similar the augmented data needs to be to the problem at hand. Our findings show that even in the controlled setting of a synthetic benchmark, achieving systematic generalization remains very difficult. After training on an augmented dataset with almost forty times more adverbs than the original problem, a non-modular baseline is not able to systematically generalize to a novel combination of a known verb and adverb. When separating the task into cognitive processes like perception and navigation, a modular neural network is able to utilize the augmented data and generalize more systematically, achieving 70% and 40% exact match increase over state-of-the-art on two gSCAN tests that have not previously been improved. We hope that this work gives insight into the drivers of systematic generalization, and what we still need to improve for neural networks to learn more like humans do.