# Compositionality and Generalization in Emergent Languages

Source: [Compositionality and Generalization in Emergent Languages](<https://www.sota2.com/research/paper/68fa675b1e209131cea15cec>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:12:13.696Z

Authors: Rahma Chaabouni, Eugene Kharitonov, Diane Bouchacourt, Emmanuel Dupoux, Marco Baroni

Published: 2020-04-20T08:30:14.000Z

## Abstract

Natural language allows us to refer to novel composite concepts by combining expressions denoting their parts according to systematic rules, a property known as \\emph{compositionality}. In this paper, we study whether the language emerging in deep multi-agent simulations possesses a similar ability to refer to novel primitive combinations, and whether it accomplishes this feat by strategies akin to human-language compositionality. Equipped with new ways to measure compositionality in emergent languages inspired by disentanglement in representation learning, we establish three main results. First, given sufficiently large input spaces, the emergent language will naturally develop the ability to refer to novel composite concepts. Second, there is no correlation between the degree of compositionality of an emergent language and its ability to generalize. Third, while compositionality is not necessary for generalization, it provides an advantage in terms of language transmission: The more compositional a language is, the more easily it will be picked up by new learners, even when the latter differ in architecture from the original agents. We conclude that compositionality does not arise from simple generalization pressure, but if an emergent language does chance upon it, it will be more likely to survive and thrive.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Emergent Communication Evaluation on ImageNet-100 20-class subset](<https://www.sota2.com/research/sota/emergent-communication-evaluation-on-imagenet-100-20-class-subset.md>): Emergent Communication Evaluation; ImageNet-100 20-class subset; Top-1 Accuracy: 24.04.

- [Emergent Message Classification on CIFAR-10 standard (test)](<https://www.sota2.com/research/sota/emergent-message-classification-on-cifar-10-standard-test.md>): Emergent Message Classification; CIFAR-10 standard (test); Top-1 Accuracy: 35.94.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2004.09124>)

[PDF](<https://arxiv.org/pdf/2004.09124.pdf>)