Research map / Multi-agent systems and games / Human-AI and goal-conditioned
VAIN: Attentional Multi-agent Predictive Modeling
Yedid Hoshen
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 |
|---|---|---|---|
| Add attentional weighting to CommNet-style agent communication | VAIN · Chess MPP | accuracy: +2.9% | Improved |
| Use per-agent outputs and attention rather than the IN formulation | VAIN · Chess MPP | accuracy: +1.8% | Improved |
| Use attention-based interactions instead of CommNet communication | VAIN · Soccer SVPP | prediction error (m): 1.04 -> 1.03 | Improved |
| Use attentional interactions instead of CommNet communication | VAIN · Bouncing Balls | RMS: 0.510 -> 0.135 | Improved |
| Use VAIN instead of IN at similar computation budgets | VAIN · Bouncing Balls | RMS: 0.139 -> 0.135 | Improved |
| Use an attention mechanism instead of unweighted communication | VAIN · Bouncing Balls | RMS | Improved |
| Add multiple hops (ablation) | VAIN · Multi-agent prediction experiments | performance | No clear effect |
Closest papers by meaning
- Interaction Modeling with Multiplex Attention (NeurIPS 2022 · similarity 0.90)
- Generative Attention Networks for Multi-Agent Behavioral Modeling (AAAI 2020 · similarity 0.89)
- Pairwise is Not Enough: Hypergraph Neural Networks for Multi-Agent Pathfinding (ICLR 2026 · similarity 0.85)
- Arena: A General Evaluation Platform and Building Toolkit for Multi-Agent Intelligence (AAAI 2020 · similarity 0.84)
- Inverse Attention Agents for Multi-Agent Systems (ICLR 2025 · similarity 0.84)
- Graph-of-Agents: A Graph-based Framework for Multi-Agent LLM Collaboration (ICLR 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.
