Reproduce “Emergent Language: A Survey and Taxonomy” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jannik Peters, Constantin Waubert de Puiseau, Hasan Tercan, Arya Gopikrishnan, Gustavo Adolpho Lucas De Carvalho, Christian Bitter, Tobias Meisen
The paper provides a systematic survey and taxonomy for discrete emergent language, synthesizing 181 publications to offer a standardized framework, categorize evaluation metrics, and identify research gaps. It aims to foster coherence and comparability in the field of multi-agent emergent communication.
Key insights
- Emergent language research predominantly features fixed message lengths, unlike the variable lengths found in natural language.
- Evaluation metrics are imbalanced, with semantic and morphological metrics being well-established, while syntax metrics are significantly underrepresented and rarely adopted.
- Defining 'optimal' emergent language systems is complex, as ideal metric values often represent a nuanced balance rather than simple minimization or maximization.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.