Learning to model other minds

Learning with opponent-learning awareness

Triadic Werewolf: A Jester Role for Multi-Hop Theory of Mind in LLMs

arXiv:2606. 27909v1 Announce Type: cross Abstract: Theory-of-mind evaluations of large language models typically use dyadic social-deduction games, where every observable cue points to a single hidden side, so a model with strong language priors can score well without ever simulating opponents' incentives.

Embedded Universal Predictive Intelligence: a coherent framework for multi-agent learning

arXiv:2511. 22226v2 Announce Type: replace Abstract: The standard theory of model-free reinforcement learning assumes that the environment dynamics are stationary and that agents are decoupled from their environment, such that policies are treated as being separate from the world they inhabit.

CCKS: Consensus-based Communication and Knowledge Sharing

arXiv:2606. 12281v1 Announce Type: cross Abstract: In Decentralized Training and Decentralized Execution (DTDE) for cooperative Multi-Agent Reinforcement Learning (MARL), action-advising-based knowledge sharing promotes interpretable and scalable cooperation among agents.

Training Small LLMs as Spatial Multi-Agent Policies

arXiv:2608. 01425v1 Announce Type: cross Abstract: Training LLM-based multi-agent systems with multi-agent reinforcement learning is rapidly gaining traction, and a parallel line of work argues that such systems should be judged by their behavior, not only their reward.

Learning Montezuma’s Revenge from a single demonstration

We’ve trained an agent to achieve a high score of 74,500 on Montezuma’s Revenge from a single human demonstration, better than any previously published result. Our algorithm is simple: the agent plays a sequence of games starting from carefully chosen states from the demonstration, and learns from them by optimizing the game score using PPO, the same reinforcement learning algorithm that underpins OpenAI Five.

Synchronizing Beliefs with Second-Order Theory-of-Mind in Human-Autonomy Teams (Extended Version)

arXiv:2608. 11229v1 Announce Type: new Abstract: Comparative feedback, asking people which of two behaviors they prefer, has become a standard way to align robot and agent behavior with human intent when the reward itself cannot be specified directly.

Playing Along: Learning a Double-Agent Defender for Belief Steering via Theory of Mind

arXiv:2604. 11666v2 Announce Type: replace-cross Abstract: As large language models (LLMs) become the engine behind conversational systems, their ability to reason about the intentions and states of their dialogue partners (i.

Quantifying generalization in reinforcement learning

We’re releasing CoinRun, a training environment which provides a metric for an agent’s ability to transfer its experience to novel situations and has already helped clarify a longstanding puzzle in reinforcement learning. CoinRun strikes a desirable balance in complexity: the environment is simpler than traditional platformer games like Sonic the Hedgehog but still poses a worthy generalization challenge for state of the art algorithms.

Social Gym and SPaRTan: Benchmarking and Improving LLM Social Reasoning via Multi-Agent Game Tournaments

arXiv:2608. 09128v1 Announce Type: cross Abstract: LLM agents are increasingly deployed in multi-agent social settings where they must cooperate, negotiate, and adapt to other agents.

Communication-Enhanced Tutoring for Efficient Decentralized Multi-Agent Reinforcement Learning

arXiv:2508. 13661v4 Announce Type: replace Abstract: Centralized Training with Decentralized Execution (CTDE) is the dominant paradigm in multi-agent reinforcement learning (MARL), enabling agents to act independently at test time while leveraging additional information during training.

A Contextual-Bandit Oversight Game with Two-Sided Informational Asymmetry

arXiv:2607. 00155v1 Announce Type: new Abstract: We study runtime human oversight of an AI agent when private information runs in both directions: the human privately knows her reward function, while the AI privately knows the quality of the action it proposes.