ICLR 2026 accepted papers on Multi-agent systems and games
256 of the 5,358 papers accepted at ICLR 2026 (4.8%) are on Multi-agent systems and games, compared with 3.2% at ICLR 2025. They fall into 4 topics.
See ICLR 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
LLM agents · 197 papers
- CellAgent: LLM-Driven Multi-Agent Framework for Natural Language-Based Single-Cell Analysis (42 citations)
- CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale (2 citations)
- Emergent Coordination in Multi-Agent Language Models (2 citations)
- CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards (1 citation)
- Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions (1 citation)
- Grounding Computer Use Agents on Human Demonstrations (1 citation)
- Scaling Generalist Data-Analytic Agents (1 citation)
- ST-WebAgentBench: A Benchmark for Evaluating Safety and Trustworthiness in Web Agents (1 citation)
- WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research (1 citation)
- Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learning
- Advancing Multi-agent Traffic Simulation via R1-Style Reinforcement Fine-Tuning
- AgenTracer: Who Is Inducing Failure in the LLM Agentic Systems?
and 185 more in the interactive map.
Human-AI and goal-conditioned · 28 papers
- Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration (1 citation)
- Efficient Agent Training for Computer Use (1 citation)
- Theory of Space: Can Foundation Models Construct Spatial Beliefs through Active Exploration? (1 citation)
- A Framework for Studying AI Agent Behavior: Evidence from Consumer Choice Experiments
- AutoLibra: Agent Metric Induction from Open-Ended Human Feedback
- Building spatial world models from sparse transitional episodic memories
- CooperTrim: Adaptive Data Selection for Uncertainty-Aware Cooperative Perception
- DecompGAIL: Learning Realistic Traffic Behaviors with Decomposed Multi-Agent Generative Adversarial Imitation Learning
- GT-Space: Enhancing Heterogeneous Collaborative Perception with Ground Truth Feature Space
- Improving Human-AI Coordination through Online Adversarial Training and Generative Models
- JointDiff: Bridging Continuous and Discrete in Multi-Agent Trajectory Generation
- Language and Experience: A Computational Model of Social Learning in Complex Tasks
and 16 more in the interactive map.
Cooperative multi-agent RL · 25 papers
- Inter-Agent Relative Representations for Multi-Agent Option Discovery
- Multi-Action Self-Improvement For Neural Combinatorial Optimization
- Multi-agent Coordination via Flow Matching
- Multi-Agent Guided Policy Optimization
- Negotiated Reasoning: On Provably Addressing Relative Over-Generalization
- Personalized Collaborative Learning with Affinity-Based Variance Reduction
- Potentially Optimal Joint Actions Recognition for Cooperative Multi-Agent Reinforcement Learning
- Retaining Suboptimal Actions to Follow Shifting Optima in Multi-Agent Reinforcement Learning
- Safe Continuous-time Multi-Agent Reinforcement Learning via Epigraph Form
- SocialJax: An Evaluation Suite for Multi-agent Reinforcement Learning in Sequential Social Dilemmas
- Triple-BERT: Do We Really Need MARL for Order Dispatch on Ride-Sharing Platforms?
- When Is Diversity Rewarded in Cooperative Multi-Agent Learning?
and 13 more in the interactive map.
Games and equilibria · 6 papers
- Look-ahead Reasoning with a Learned Model in Imperfect Information Games
- Matching multiple experts: on the exploitability of multi-agent imitation learning
- Risk-Sensitive Agent Compositions
- SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning
- Towards Sustainable Investment Policies Informed by Opponent Shaping
- Incentives in Federated Learning with Heterogeneous Agents