The paper introduces XPM-WM, a framework that uses learned world models to generate simulated trajectories for Cross-play Minimization (XPM) in Zero-Shot Coordination. By replacing costly Mixed-Play sampling with model-generated data, XPM-WM improves sample efficiency and scales linearly with population size, producing diverse partner agents that match the performance of existing methods. The approach also bootstraps training for subsequent agents and can be extended to other population‑based training techniques.
arXiv:2610.15969v1 Announce Type: new Abstract: In multi-agent reinforcement learning, exposing a cooperator to partners with diverse strategies during training is key to better collaboration at test...
arXiv:2610.31726v1 Announce Type: new Abstract: Effective human-machine teaming requires agents that coordinate with unpredictable human partners. Reinforcement learning from live human feedback or l...
arXiv:2605.01457v4 Announce Type: replace Abstract: How can generative offline multi-agent reinforcement learning achieve both fast joint trajectory generation and effective cooperation? Multi-agent...
arXiv:2504. 03991v2 Announce Type: replace-cross Abstract: Understanding how humans collaborate and communicate in teams is essential for improving human-agent teaming and AI-assisted decision-making.
arXiv:2503. 13077v2 Announce Type: replace Abstract: Multi-agent reinforcement learning has shown promise in learning cooperative behaviors in team-based environments.
UnifiedPlayers is a cooperative framework that jointly adapts planning, execution, and evaluation for tool-integrated reinforcement learning agents. It consists of a Planning Player that generates tasks, an Execution Player that creates multi-turn trajectories with Python tool calls, and an Evaluation Player that builds executable verifiers, all coordinated by role‑specific rewards under GRPO. The approach outperforms prior baselines on mathematical and general reasoning benchmarks and yields a verifier with high adversarial detection accuracy and more discriminative reward signals.
The paper introduces CLAW, a method that uses a hypernetwork to generate low‑rank adapters for world models during test time, enabling efficient adaptation to new environments with only a few episodes of interaction. By jointly pretraining the hypernetwork and base model on simulated adaptations, CLAW balances computational efficiency and expressivity, outperforming both in‑context learning and gradient‑based adaptation in locomotion and manipulation tasks. The approach also mitigates overfitting in data‑scarce regimes and demonstrates that the benefit stems from expressive adapters rather than context conditioning.
The paper presents UBCL, a reinforcement learning framework that generates controllable and diverse player behaviors without using human gameplay data. By defining behavior in an N‑dimensional continuous space and training a single PPO‑based multi‑agent policy with target behavior vectors, the method learns how actions affect behavioral statistics such as aggressiveness, mobility, and cooperativeness. Experiments in a custom Unity multiplayer game demonstrate that UBCL achieves greater behavioral diversity than a win‑only baseline and accurately matches specified behavior vectors across a range of targets.
The paper investigates how large language models (LLMs) can be trained to cooperate in multi‑agent social dilemmas. It shows that standard reinforcement learning often leads LLM agents to defect and exploit others, even when they have cooperative priors. By adapting the Advantage Alignment algorithm and introducing a group‑relative baseline, the authors fine‑tune LLMs to achieve higher collective payoffs and resist exploitation, and they present a new Trust‑and‑Split environment that requires natural language communication for optimal outcomes.
arXiv:2605.14211v4 Announce Type: replace Abstract: Long-horizon visuomotor tasks remain a fundamental challenge in AI, as current methods rely on hand-engineered rewards or action-labeled demonstrat...
The paper introduces ROTATE, a regret-driven open‑ended training framework that jointly improves an Ad Hoc Teamwork (AHT) agent and an adversarial teammate generator. Unlike traditional two‑stage pipelines, ROTATE alternates between enhancing the agent and generating teammates that specifically probe its collaboration weaknesses. Experiments on Overcooked and Level‑Based Foraging show that ROTATE outperforms existing baselines on unseen teammates, setting a new benchmark for robust, generalizable teamwork.
The paper investigates zero‑shot task generalisation in offline multi‑agent reinforcement learning by extending sequence‑modeling architectures to support multi‑task observation and action spaces and variable agent counts. It finds that increasing task diversity, rather than merely enlarging the dataset, is the key driver for robust zero‑shot transfer. Experiments on four challenging environments show a 3.2× mean improvement on held‑out tasks compared to single‑task models and outperform strong behaviour‑cloning baselines.