Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Generative Attention Networks for Multi-Agent Behavioral Modeling
Guangyu Li, Bo Jiang, Hao Zhu, Zhengping Che, Yan Liu
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 |
|---|---|---|---|
| add attention-based interaction modeling (GAMAN vs GAMAN-NI) (ablation) | GAMAN · Spring-Ball | L2 distance: 0.2624 -> 0.1728 | Improved |
| add attention-based interaction modeling (GAMAN vs GAMAN-NI) (ablation) | GAMAN · Stanford Drone | L2 distance: 0.0264 -> 0.0203 | Improved |
| add attention-based interaction modeling (GAMAN vs GAMAN-NI) (ablation) | GAMAN · NBA | L2 distance: 0.2465 -> 0.1720 | Improved |
| add attention-based interaction modeling (GAMAN vs GAMAN-NI) (ablation) | GAMAN · Stanford Drone | MR0.5: 0.0052 -> 0.0047 | Mixed |
| add attention-based interaction modeling (GAMAN vs GAMAN-NI) (ablation) | GAMAN · Spring-Ball | log-likelihood: -0.75 -> 1.21 | Improved |
| add attention-based interaction modeling (GAMAN vs GAMAN-NI) (ablation) | GAMAN · Stanford Drone | log-likelihood: 4.04 -> 6.12 | Improved |
| add attention-based interaction modeling (GAMAN vs GAMAN-NI) (ablation) | GAMAN · NBA | log-likelihood: -1.1 -> 0.80 | Improved |
| GAMAN vs best deep baseline (NRI) on NBA maxL2/L2 | GAMAN · NBA | maxL2: 0.3647 (NRI) vs 0.3794 (GAMAN) | Hurt |
| GAMAN vs Velocity baseline on Stanford Drone MR0.5 | GAMAN · Stanford Drone | MR0.5: 0.0031 (IN) vs 0.0047 (GAMAN) | Hurt |
Closest papers by meaning
- Grounded Answers for Multi-agent Decision-making Problem through Generative World Model (NeurIPS 2024 · similarity 0.86)
- Inverse Attention Agents for Multi-Agent Systems (ICLR 2025 · similarity 0.86)
- G-Memory: Tracing Hierarchical Memory for Multi-Agent Systems (NeurIPS 2025 · similarity 0.86)
- Learning to Cooperate with Humans using Generative Agents (NeurIPS 2024 · similarity 0.86)
- APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplay (NeurIPS 2025 · similarity 0.85)
- Mental-Models for Multi-Agent Systems (NeurIPS 2026 · similarity 0.85)
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.
