本研究探讨大型语言模型（LLMs）对高风险心理健康群体的无故攻击，提出三项贡献：评估攻击、研究偏见传播框架及污名化程度。结果显示，心理健康实体在攻击叙事中占中心，强调减轻偏见的必要性。

本研究探讨了紧急沟通（EC）的解读及其与自然语言（NL）的关系，采用无监督神经机器翻译技术，发现任务复杂性和语义多样性影响EC的可翻译性。尽管复杂任务使EC难以解释，但仍适合翻译。

基于大型语言模型的自主代理人在任务的广泛范围内实现了显著的改进。我们提出了一个多代理人框架，通过实验证明该框架能够有效地部署优于单个代理人的多代理人群体。我们研究了群体中个体代理人之间的社交行为，并讨论了一些利用积极行为和缓解消极行为以提高多代理人群体的协作潜力的可能策略。

Peer-to-peer systems treat each cluster node as equal; there is no strict leader. This means there is no explicit leader election process as happens in the Leader and Followers...

In our recent paper, we explore how populations of deep reinforcement learning (deep RL) agents can learn microeconomic behaviours, such as production, consumption, and trading of goods. We find...

As you all know, the simulation of universe 2813/9301 is now coming to a close. This simulation is notable for being the first simulated universe suitable for hosting intelligent life, but...

As you all know, the simulation of universe 2813/9301 is now coming to a close. This simulation is notable for being the first simulated universe suitable for hosting intelligent life, but...