# Compositional Generalization in Unsupervised Compositional Representation Learning: A Study on Disentanglement and Emergent Language

arXiv: https://arxiv.org/abs/2210.00482 · HTML: https://mlsanity.com/paper/2210.00482/

## Key facts

- Cited by 42 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 55 references that are in the map.
- First posted on arXiv 2 Oct 2022.

## Builds on (most cited first, 30 of 34)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables](https://mlsanity.com/paper/1611.00712/) (2016)
- [Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations](https://mlsanity.com/paper/1811.12359/) (2018)
- [Disentangling by Factorising](https://mlsanity.com/paper/1802.05983/) (2018)
- [Winoground: Probing Vision and Language Models for Visio-Linguistic Compositionality](https://mlsanity.com/paper/2204.03162/) (2022)
- [Semi-Supervised Learning with Deep Generative Models](https://mlsanity.com/paper/1406.5298/) (2014)
- [Understanding disentangling in β-VAE](https://mlsanity.com/paper/1804.03599/) (2018)
- [Variational Inference of Disentangled Latent Concepts from Unlabeled Observations](https://mlsanity.com/paper/1711.00848/) (2017)
- [Deep Learning: A Critical Appraisal](https://mlsanity.com/paper/1801.00631/) (2018)
- [SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decomposition](https://mlsanity.com/paper/2001.02407/) (2020)
- [Isolating Sources of Disentanglement in Variational Autoencoders](https://mlsanity.com/paper/1802.04942/) (2018)
- [Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols](https://mlsanity.com/paper/1705.11192/) (2017)
- [Task-Driven Modular Networks for Zero-Shot Compositional Learning](https://mlsanity.com/paper/1905.05908/) (2019)
- [Robustly Disentangled Causal Mechanisms: Validating Deep Representations for Interventional Robustness](https://mlsanity.com/paper/1811.00007/) (2018)
- [Compositionality and Generalization in Emergent Languages](https://mlsanity.com/paper/2004.09124/) (2020)
- [On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Dataset](https://mlsanity.com/paper/1906.03292/) (2019)
- [Building machines that adapt and compute like brains](https://mlsanity.com/paper/1711.04203/) (2017)
- [Measuring Compositionality in Representation Learning](https://mlsanity.com/paper/1902.07181/) (2019)
- [Bias and Generalization in Deep Generative Models: An Empirical Study](https://mlsanity.com/paper/1811.03259/) (2018)
- [Compositional Languages Emerge in a Neural Iterated Learning Model](https://mlsanity.com/paper/2002.01365/) (2020)
- [On the Transfer of Disentangled Representations in Realistic Settings](https://mlsanity.com/paper/2010.14407/) (2020)
- [Structured Disentangled Representations](https://mlsanity.com/paper/1804.02086/) (2018)
- [Visual Representation Learning Does Not Generalize Strongly Within the Same Domain](https://mlsanity.com/paper/2107.08221/) (2021)
- [Learning Compositional Representations for Few-Shot Recognition](https://mlsanity.com/paper/1812.09213/) (2018)

## Built on by (most cited first, 30 of 42)

- [ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty](https://mlsanity.com/paper/2408.14339/) (2024)
- [Efficient Data Collection for Robotic Manipulation via Compositional Generalization](https://mlsanity.com/paper/2403.05110/) (2024)
- [On Provable Length and Compositional Generalization](https://mlsanity.com/paper/2402.04875/) (2024)
- [When does compositional structure yield compositional generalization? A kernel theory](https://mlsanity.com/paper/2405.16391/) (2024)
- [A Review of the Applications of Deep Learning-Based Emergent Communication](https://mlsanity.com/paper/2407.03302/) (2024)
- [Can Models Learn Skill Composition from Examples?](https://mlsanity.com/paper/2409.19808/) (2024)
- [Improving Compositional Generalization Using Iterated Learning and Simplicial Embeddings](https://mlsanity.com/paper/2310.18777/) (2023)
- [What makes a language easy to deep-learn? Deep neural networks and humans similarly benefit from compositional structure](https://mlsanity.com/paper/2302.12239/) (2023)
- [Deciphering the Role of Representation Disentanglement: Investigating Compositional Generalization in CLIP Models](https://mlsanity.com/paper/2407.05897/) (2024)
- [How Diffusion Models Learn to Factorize and Compose](https://mlsanity.com/paper/2408.13256/) (2024)
- [A Survey on Compositional Generalization in Applications](https://mlsanity.com/paper/2302.01067/) (2023)
- [Lewis's Signaling Game as beta-VAE For Natural Word Lengths and Segments](https://mlsanity.com/paper/2311.04453/) (2023)
- [Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models](https://mlsanity.com/paper/2406.16135/) (2024)
- [When and How Does CLIP Enable Domain and Compositional Generalization?](https://mlsanity.com/paper/2502.09507/) (2025)
- [Towards Compositionality in Concept Learning](https://mlsanity.com/paper/2406.18534/) (2024)
- [The Curious Case of Representational Alignment: Unravelling Visio-Linguistic Tasks in Emergent Communication](https://mlsanity.com/paper/2407.17960/) (2024)
- [Vector-based Representation is the Key: A Study on Disentanglement and Compositional Generalization](https://mlsanity.com/paper/2305.18063/) (2023)
- [Imagine the Unseen World: A Benchmark for Systematic Generalization in Visual World Models](https://mlsanity.com/paper/2311.09064/) (2023)
- [Do Diffusion Models Learn Semantically Meaningful and Efficient Representations?](https://mlsanity.com/paper/2402.03305/) (2024)
- [Compositional Generalization via Forced Rendering of Disentangled Latents](https://mlsanity.com/paper/2501.18797/) (2025)
- [Probabilistic and Semantic Descriptions of Image Manifolds and Their Applications](https://mlsanity.com/paper/2307.02881/) (2023)
- [Unsupervised Translation of Emergent Communication](https://mlsanity.com/paper/2502.07552/) (2025)
- [Correcting Flaws in Common Disentanglement Metrics](https://mlsanity.com/paper/2304.02335/) (2023)
- [Enriching Disentanglement: From Logical Definitions to Quantitative Metrics](https://mlsanity.com/paper/2305.11512/) (2023)
- [SurroCBM: Concept Bottleneck Surrogate Models for Generative Post-hoc Explanation](https://mlsanity.com/paper/2310.07698/) (2023)
- [Inducing Systematicity in Transformers by Attending to Structurally Quantized Embeddings](https://mlsanity.com/paper/2402.06492/) (2024)
- [Controllable Image Generation with Composed Parallel Token Prediction](https://mlsanity.com/paper/2405.06535/) (2024)
- [SimSiam Naming Game: A Unified Approach for Emergent Communication and Representation Learning](https://mlsanity.com/paper/2410.21803/) (2024)
- [Exploring Compositional Generalization of Multimodal LLMs for Medical Imaging](https://mlsanity.com/paper/2412.20070/) (2024)
- [Randomly Sampled Language Reasoning Problems Elucidate Limitations of In-Context Learning](https://mlsanity.com/paper/2501.02825/) (2025)

## Abstract (arXiv)

Deep learning models struggle with compositional generalization, i.e. the ability to recognize or generate novel combinations of observed elementary concepts. In hopes of enabling compositional generalization, various unsupervised learning algorithms have been proposed with inductive biases that aim to induce compositional structure in learned representations (e.g. disentangled representation and emergent language learning). In this work, we evaluate these unsupervised learning algorithms in terms of how well they enable compositional generalization. Specifically, our evaluation protocol focuses on whether or not it is easy to train a simple model on top of the learned representation that generalizes to new combinations of compositional factors. We systematically study three unsupervised representation learning algorithms - β-VAE, β-TCVAE, and emergent language (EL) autoencoders - on two datasets that allow directly testing compositional generalization. We find that directly using the bottleneck representation with simple models and few labels may lead to worse generalization than using representations from layers before or after the learned representation itself. In addition, we find that the previously proposed metrics for evaluating the levels of compositionality are not correlated with actual compositional generalization in our framework. Surprisingly, we find that increasing pressure to produce a disentangled representation produces representations with worse generalization, while representations from EL models show strong compositional generalization. Taken together, our results shed new light on the compositional generalization behavior of different unsupervised learning algorithms with a new setting to rigorously test this behavior, and suggest the potential benefits of delevoping EL learning algorithms for more generalizable representations.