Research map / Multi-agent systems and games / Games and equilibria
Cooperation and social dilemmas
Studies how learning agents cooperate or collude in dilemmas and sustain cooperation mechanisms.
95 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Games and equilibria, keyword cluster “game · player · sum”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
18 papers in 2023–24 and 32 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 14.2% to 22.4%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Incentives for Strategic Behavior in Fisher Market Games (AAAI 2016 · 11 citations)
- 2017 · Multi-View Decision Processes: The Helper-AI Problem (NeurIPS 2017 · 17 citations)
- 2018 · Inequity aversion improves cooperation in intertemporal social dilemmas (NeurIPS 2018 · 71 citations)
- 2019 · Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning (ICLR 2019 · 48 citations)
- 2020 · Contextual Games: Multi-Agent Learning with Side Information (NeurIPS 2020 · 6 citations)
- 2021 · Power in Liquid Democracy (AAAI 2021 · 14 citations)
- 2022 · Inverse Game Theory for Stackelberg Games: the Blessing of Bounded Rationality (NeurIPS 2022 · 6 citations)
- 2023 · Emergence of Punishment in Social Dilemma with Environmental Feedback (AAAI 2023 · 30 citations)
- 2024 · Abstraction of Situation Calculus Concurrent Game Structures (AAAI 2024 · 3 citations)
- 2025 · A Unified Model of Direct and Indirect Reciprocity in Multichannel Games (AAAI 2025 · 2 citations)
- 2026 · CoopEval: Benchmarking Cooperation-Sustaining Mechanisms and LLM Agents in Social Dilemmas (ICML 2026 · 0 citations)
Earliest papers here
- Incentives for Strategic Behavior in Fisher Market Games (AAAI 2016 · 11 citations)
- Sequence-Form and Evolutionary Dynamics: Realization Equivalence to Agent Form and Logit Dynamics (AAAI 2016 · 5 citations)
- Multi-View Decision Processes: The Helper-AI Problem (NeurIPS 2017 · 17 citations)
What has been tried
63 papers here report 476 measured changes (a component added, removed or swapped, compared with a baseline): helped 63%, no clear effect 8%, hurt 23%, mixed 6%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add assumed embedding noise to the Bayesian spymaster | Codenames, deterministic environment, in-distribution guessers · win rate | 0.960 -> 0.860 | Hurt | Improving Cooperation in Language Games with Bayesian Inference and the Cognitive Hierarchy (AAAI 2025 · 1 citation) |
| Remove task-specific reward R2 | MATH — LLaMA3.1 70B · accuracy | 81.47 -> 77.55 | Hurt | From Debate to Equilibrium: Belief‑Driven Multi‑Agent LLM Reasoning via Bayesian Nash Equilibrium (ICML 2025 · 1 citation) |
| add sign alignment to AgA | toy two-player differentiable mixed-motive game · steps to complete trajectory | approximately 15% | Helped | Aligning Individual and Collective Objectives in Multi-Agent Cooperation (NeurIPS 2024 · 1 citation) |
| replace inferred co-player policy with a uniform policy | Cleanup · convergence speed and performance | Hurt | Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games (NeurIPS 2024 · 1 citation) | |
| form multiple joint federations dynamically instead of fixed disjoint federations | Matthew Effect · collective returns | Hurt | Beyond Mandatory Federations: Balancing Egoism, Utilitarianism and Egalitarianism in Mixed-Motive Games (AAAI 2025 · 0 citations) | |
| replace multi-agent gameplay with self-play | 9-player Seer-Witch-Guard · average win rate | 0.534 -> 0.483 | Hurt | Multi-agent KTO: Enhancing Strategic Interactions of Large Language Model in Language Game (NeurIPS 2025 · 0 citations) |
| Use fully coordinated teams rather than groups of two or independent agents | Two-team ZSPTG with four agents per team · convergence speed | Helped | Team-Fictitious Play for Reaching Team-Nash Equilibrium in Multi-team Games (NeurIPS 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 24 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- GSM8K · 2 of 24 papers
Baselines
- IPPO · 2 of 24 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 50 papers; from OpenAlex and the papers' affiliations).
- Carnegie Mellon University · 6 papers
- University of Oxford · 5 papers
- University of California, Berkeley · 4 papers
- Google DeepMind · 3 papers
- Massachusetts Institute of Technology · 3 papers
- Northwestern Polytechnical University · 2 papers
- University of Texas at Austin · 2 papers
- Peking University · 2 papers
- Tsinghua University · 2 papers
- Fudan University · 2 papers
Use it
Download BibTeX of 40 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Learning in Markov games (111 papers)
- Opponent modeling in competitive games (75 papers)
- Mean-field game learning (21 papers)
Elsewhere in the topic
- Fair division and coalition formation (99 papers)
- Learning with strategic agents (74 papers)
- Strategic classification (51 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).