Research map / Large language models / Language, safety and interpretability
Expressivity of Emergent Languages is a Trade-off between Contextual Complexity and Unpredictability
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Increase candidate-set size while holding context fixed | Referential-game language training · Transfer across referential games | generalisation accuracy | Helped |
| Use varying rather than fixed contexts | Referential-game language training · All target games | generalisation accuracy | Helped |
| Use a moderate candidate-set size, such as 1,000, 2,500, or 5,000 | Referential-game language training · Transfer across referential games | generalisation accuracy | Helped |
| Use contrastive loss instead of conventional referential loss | Referential-game agents · Message type counts in refer10 and refer100 | number of message types: less than 500; roughly 5,000 | Helped |
| Use contrastive loss instead of conventional referential loss | Referential-game agents · Training stability | variance | Helped |
| Rely on message-type count alone to determine language expressivity | Emergent languages · Transfer across referential games | generalisation accuracy | No clear effect |
| Use mutual information as a complete measure of expressivity | Emergent languages · Transfer across referential games | generalisation accuracy | No clear effect |
Closest papers by meaning
- Lewis's Signaling Game as beta-VAE For Natural Word Lengths and Segments (ICLR 2024 · similarity 0.91)
- Understanding Emergent Abilities of Language Models from the Loss Perspective (NeurIPS 2024 · similarity 0.89)
- Unsupervised Translation of Emergent Communication (AAAI 2025 · similarity 0.89)
- A compressive-expressive communication framework for compositional representations (NeurIPS 2025 · similarity 0.88)
- AlphaZero Neural Scaling and Zipf's Law: a Tale of Board Games and Power Laws (NeurIPS 2025 · similarity 0.87)
- Bridging semantics and pragmatics in information-theoretic emergent communication (NeurIPS 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.