arXiv:2603. 05789v5 Announce Type: replace-cross Abstract: Repeated multi-agent interactions require evaluation metrics that capture not only payoff distributions but also their temporal organization.
arXiv:2605. 14879v2 Announce Type: replace-cross Abstract: Many intelligent computing and autonomous systems rely on multiple independent, often learning, agents repeatedly sharing a limited resource.
arXiv:2607. 04470v1 Announce Type: cross Abstract: Large Language Models (LLMs) offer a natural interface for translating human objectives into reward signals for cooperative multi-agent reinforcement learning (MARL), yet the training-time dynamics of this integration remain poorly understood.
The paper investigates the problem of sharing a single critic across multiple parallel environments in reinforcement learning. It shows that when environments assign different expected returns to the same state, a shared critic must reconcile conflicting value targets, which can distort advantage estimates and misguide policy updates. The authors propose a simple fix—providing the critic with the environment index—demonstrating through bandit models and experiments on CartPole, MuJoCo, BipedalWalker, and 16 Procgen games that this conditional critic stabilizes learning and boosts returns, achieving a 40.8% improvement in aggregate normalized return on unseen levels.
The paper introduces CURB, a reward‑shaping framework that penalizes the total variation distance between an agent’s action distributions under cooperation and defection histories, thereby preventing collusive equilibria in repeated games. By linking empirical Q‑learning collusion to Simple Penal Codes, the authors prove that any non‑trivial SPC can be neutralized, and demonstrate CURB’s effectiveness in both tabular and deep Q‑learning settings for Bertrand and Cournot competition.
arXiv:2606. 11284v1 Announce Type: cross Abstract: Real-world multi-agent systems, from traffic coordination to resource allocation, are often modeled as general-sum games where individual incentives conflict with collective welfare.
arXiv:2610.00619v1 Announce Type: cross Abstract: In this paper, we extend earlier findings of supra-competitive outcomes in optimal-execution games by identifying a learned punitive mechanism that d...
FairTest is a search-based testing framework designed to uncover fairness failures in Multi-Agent Reinforcement Learning (MARL) systems. It guides candidate generation using three fitness functions—measuring observed fairness, predicting fairness from abstract states, and assessing policy decision uncertainty—and prioritizes tests based on predicted fairness and uncertainty. Evaluations on three environments and two MARL algorithms show FairTest detects significantly more fairness failures than three baselines, with a 221% increase in failure count and 23% better coverage on average.
arXiv:2601. 17454v2 Announce Type: replace-cross Abstract: Centralized value learning underlies a broad class of multi-agent reinforcement learning methods, but its claimed advantage is typically evaluated in settings that confound coordination structure with function approximation and partial observability.
Reinforcement learning (RL) fine-tuning is widely used in language model training to improve model performance on a target task while limiting drift from a reference policy. A standard way to balance this trade-off is via a KL-regularized RL objective, although this formulation does not by itself provide a principled way to set the regularization coefficient.
arXiv:2607. 05904v1 Announce Type: new Abstract: Training a language model against its own reference-free judgments (the premise of self-rewarding, self-play, and LLM-as-a-judge pipelines) assumes a model's verdict on a shown answer tracks correctness.
arXiv:2607. 23772v1 Announce Type: cross Abstract: We study a restless multi-armed bandit (RMAB) problem for a stochastic deadline scheduling application.
arXiv:2607. 26358v1 Announce Type: new Abstract: Reinforcement learning (RL) fine-tuning is widely used in language model training to improve model performance on a target task while limiting drift from a reference policy.