Research map / Probabilistic and scientific modeling / Molecules and proteins
Common Ground in Cooperative Communication
Xiaoran Hao, Yash Jhaveri, Patrick Shafto
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 |
|---|---|---|---|
| use sparse initialization (p=0.5) instead of uniform initialization (p=0.0625) | MLP-based common ground, eps=1, delta=0 · Fully discrete, varying m | cooperative index | Improved |
| run with eps=1 and several random initializations | MLP-based common ground · Fully discrete, varying m | pairwise l1-distance between minimizers | Hurt |
| increase eps from 0 toward 1 | MLP-based common ground · Fully discrete, m=20 | l1-distance between teacher and learner plans | Improved |
| increase eps from 0 toward 1 | MLP-based common ground · Fully discrete, m=20 | cooperative index | Improved |
| increase initialization range p from 0.01 to 0.8 | Variational cooperative communication, 3-Gaussian g, eps=1, delta=10 · Semi-continuous | average reconstruction loss | Hurt |
| increase initialization range p from 0.01 to 0.8 | Variational cooperative communication, 3-Gaussian g, eps=1, delta=10 · Semi-continuous | data activation percentage | Hurt |
| vary eps across values around 1 | Variational cooperative communication · Semi-continuous, 1/3/5 Gaussians | average reconstruction loss | Mixed |
| vary delta | Variational cooperative communication · Semi-continuous, 1/3/5 Gaussians | data activation percentage | No clear effect |
| vary delta | Variational cooperative communication · Semi-continuous, 1/3/5 Gaussians | average reconstruction loss | No clear effect |
| increase number of Gaussians in teacher prior g from 1 to 5 | Variational cooperative communication · Semi-continuous | optimal eps location | Mixed |
Closest papers by meaning
- Learning Multi-Agent Communication from Graph Modeling Perspective (ICLR 2024 · similarity 0.84)
- Language Grounded Multi-agent Reinforcement Learning with Human-interpretable Communication (NeurIPS 2024 · similarity 0.84)
- Semantics and Spatiality of Emergent Communication (NeurIPS 2024 · similarity 0.84)
- Socialized Coevolution: Advancing a Better World through Cross-Task Collaboration (ICML 2025 · similarity 0.84)
- More Capable, Less Cooperative? When LLMs Fail at Zero-Cost Collaboration (ICML 2026 · similarity 0.84)
- COOP$^2$: Defining, Observing, and Repairing Cooperation in LLM Multi-Agent Systems (NeurIPS 2026 · similarity 0.83)
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.
