Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Modeling Human Ad Hoc Coordination
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 |
|---|---|---|---|
| Use probability matching on maximal perceived common p-belief | Human coordination model · Thomas experiments (four knowledge conditions) | mean-squared error | Helped |
| Use probability matching on maximal perceived common p-belief rather than bounded-depth iterated matching | Human coordination model · Thomas experiments, secondary and tertiary knowledge conditions | coordination probability | Helped |
| Assume utility maximization with common p-belief | Rational p-belief model · Thomas experiments | prediction fit | Hurt |
| Assume bounded-depth iterated maximization | Iterated maximization model · Thomas experiments | prediction fit | Hurt |
| Model the human companion with matched p-belief and choose the agent's best response | Human-agent coordination strategy · Human-agent coordination simulations | marginal value | Helped |
| Use the cognitive strategy at very high risk levels | Human-agent coordination strategy · Human-agent coordination simulations, very high risk | marginal value | Hurt |
| Use the private heuristic rather than always playing the safe action | Human-agent coordination strategy · Human-agent coordination simulations, low risk | marginal value | Helped |
| Use the pair heuristic rather than always playing the safe action | Human-agent coordination strategy · Human-agent coordination simulations, high risk | marginal value | Helped |
| Set the world-model parameter δ to 0.25 rather than 0.5 | Generative model of the Thomas experiments · Thomas experiments | empirical fit: 0.25; 0.5 | Helped |
| Interpret the private-knowledge message as communicating an intention not to visit | Generative model of the private-knowledge condition · Thomas experiments, private knowledge | empirical fit | Helped |
Closest papers by meaning
- Adaptive Theory of Mind for LLM-based Multi-Agent Coordination (AAAI 2026 · similarity 0.87)
- Risk-Sensitive Theory of Mind: Coordinating with Agents of Unknown Bias using Cumulative Prospect Theory (ICML 2025 · similarity 0.87)
- Beyond Single Stationary Policies: Meta-Task Players as Naturally Superior Collaborators (NeurIPS 2024 · similarity 0.86)
- Narrowing the Collaboration Gap, Probably (NeurIPS 2026 · similarity 0.86)
- Emergent Coordination in Multi-Agent Language Models (ICLR 2026 · similarity 0.86)
- 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.