Emergent Language: A Survey and Taxonomy
arXiv:2409.02645 · doi:10.1007/s10458-025-09691-y
Abstract
The field of emergent language represents a novel area of research within the domain of artificial intelligence, particularly within the context of multi-agent reinforcement learning. Although the concept of studying language emergence is not new, early approaches were primarily concerned with explaining human language formation, with little consideration given to its potential utility for artificial agents. In contrast, studies based on reinforcement learning aim to develop communicative capabilities in agents that are comparable to or even superior to human language. Thus, they extend beyond the learned statistical representations that are common in natural language processing research. This gives rise to a number of fundamental questions, from the prerequisites for language emergence to the criteria for measuring its success. This paper addresses these questions by providing a comprehensive review of 181 scientific publications on emergent language in artificial intelligence. Its objective is to serve as a reference for researchers interested in or proficient in the field. Consequently, the main contributions are the definition and overview of the prevailing terminology, the analysis of existing evaluation methods and metrics, and the description of the identified research gaps.
published in Journal of Autonomous Agents and Multi-Agent Systems
References in corpus (45)
- A Survey on Large Language Model based Autonomous Agents
- Reflexion: Language Agents with Verbal Reinforcement Learning
- Multi-Agent Cooperation and the Emergence of (Natural) Language
- Cognitive Architectures for Language Agents
- Compositional Languages Emerge in a Neural Iterated Learning Model
- Biases for Emergent Communication in Multi-agent Reinforcement Learning
- Developmentally motivated emergence of compositional communication via template transfer
- Measuring Compositionality in Representation Learning
- The Emergence of Compositional Languages for Numeric Concepts Through Iterated Learning in Neural Agents
- Learning to Ground Multi-Agent Communication with Autoencoders
- Shaping representations through communication: community size effect in artificial learning systems
- Morphology Matters: A Multilingual Language Modeling Analysis
- Towards More Human-like AI Communication: A Review of Emergent Communication Research
- Towards Human-Agent Communication via the Information Bottleneck Principle
- Emergence of Theory of Mind Collaboration in Multiagent Systems
- Networked Multi-Agent Reinforcement Learning with Emergent Communication
- Interpretable agent communication from scratch (with a generic visual processor emerging on the side)
- Interpretable Learned Emergent Communication for Human-Agent Teams
- Do Embodied Agents Dream of Pixelated Sheep: Embodied Decision Making using Language Guided World Modelling
- Measuring non-trivial compositionality in emergent communication
- Linking Emergent and Natural Languages via Corpus Transfer
- Emergent Discrete Communication in Semantic Spaces
- Establishing linguistic conventions in task-oriented primeval dialogue
- ESCELL: Emergent Symbolic Cellular Language
- Reinforcement Communication Learning in Different Social Network Structures
- Learning to Draw: Emergent Communication through Sketching
- On the role of population heterogeneity in emergent communication
- Re-conceptualising the Language Game Paradigm in the Framework of Multi-Agent Reinforcement Learning
- A Practical Guide to Studying Emergent Communication through Grounded Language Games
- Inductive Bias and Language Expressivity in Emergent Communication
- Avoiding hashing and encouraging visual semantics in referential emergent language games
- Analyzing Language Learned by an Active Question Answering Agent
- Learning to Ground Decentralized Multi-Agent Communication with Contrastive Learning
- An Analysis of Discretization Methods for Communication Learning with Multi-Agent Reinforcement Learning
- Emergent Communication: Generalization and Overfitting in Lewis Games
- Compositional Generalization in Unsupervised Compositional Representation Learning: A Study on Disentanglement and Emergent Language
- Emergence of Numeric Concepts in Multi-Agent Autonomous Communication
- A perspective on multi-agent communication for information fusion
- Lewis's Signaling Game as beta-VAE For Natural Word Lengths and Segments
- On (Emergent) Systematic Generalisation and Compositionality in Visual Referential Games with Straight-Through Gumbel-Softmax Estimator
- Biology and Compositionality: Empirical Considerations for Emergent-Communication Protocols
- Modeling Emergent Lexicon Formation with a Self-Reinforcing Stochastic Process
- Emergent Communication in Interactive Sketch Question Answering
- Inductive Bias for Emergent Communication in a Continuous Setting
- Learning to Communicate with Strangers via Channel Randomisation Methods