Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Emergent Communication under Varying Sizes and Connectivities
Jooyeon Kim, Alice Oh
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| optimize link structure with CMA-ES instead of random links | communication graph · Hard constraint, N=16, 3 outgoing links | success rate: 0.578 -> 0.894 | Improved |
| optimize links instead of neighbor-to-neighbor links | communication graph · Hard constraint, N=16, 3 outgoing links | success rate: 0.454 -> 0.894 | Improved |
| optimize links instead of all-to-one links | communication graph · Hard constraint, N=16, 3 outgoing links | success rate: 0.384 -> 0.894 | Improved |
| restrict to 3 outgoing links instead of all-to-all | communication graph · Hard constraint, N=16, 3 outgoing links | success rate: 0.950 -> 0.894 | Mixed |
| add any communication links versus none | communication graph · Hard constraint, N=16, 3 outgoing links | success rate: 0.236 -> 0.894 | Improved |
| optimize with distance penalty instead of random links at similar penalty | communication graph · Soft constraint, N=16, k=0.0012 | success rate: 0.513 -> 0.877 | Improved |
| optimize with distance penalty instead of neighbor-to-neighbor | communication graph · Soft constraint, N=16, k=0.0012 | success rate: 0.454 -> 0.877 | Improved |
| optimize instead of 1-to-All graph | communication graph · Soft constraint, N=16, k=0.0012 | success rate: 0.251 -> 0.877 | Improved |
| increase group size N up to hundreds | emergent language · bipartite referential game | message agreement (cosine similarity) | Hurt |
| increase group size N up to hundreds | emergent language · bipartite referential game | communication success rate | No clear effect |
| increase max communication range j in neighbor-to-neighbor graph | emergent language · N=64 | success rate between non-neighbors | Improved |
| increase max communication range j from 1 to 16 | emergent language · N=64 | dialect structure by locality: j = 1 -> j = 16 | Mixed |
Closest papers by meaning
- Scaling Large Language Model-based Multi-Agent Collaboration (ICLR 2025 · similarity 0.89)
- Lewis's Signaling Game as beta-VAE For Natural Word Lengths and Segments (ICLR 2024 · similarity 0.88)
- Emergent Coordination in Multi-Agent Language Models (ICLR 2026 · similarity 0.88)
- Semantics and Spatiality of Emergent Communication (NeurIPS 2024 · similarity 0.87)
- A compressive-expressive communication framework for compositional representations (NeurIPS 2025 · similarity 0.87)
- More Capable, Less Cooperative? When LLMs Fail at Zero-Cost Collaboration (ICML 2026 · 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.
