我们介绍结构化主动推理（structured active inference），这是运用范畴系统理论工具对主动推理进行的一次大规模推广与形式化。我们在形式上将生成模型刻画为“位于接口上”的系统；该接口是对通常的 Markov blanket（马尔可夫毯）概念的组合式抽象。于是，智能体成为其生成模型的“控制器”，并在形式上与生成模型对偶。这为主动推理的研究图景开启了新的前景，例如：具有结构化接口的智能体（包括具有“模式依赖性”的智能体，或与计算机 API 交互的智能体）；能够管理其他智能体的智能体；以及利用主动推理改变其（内部或外部）结构的“元智能体”。借助结构化接口，我们还获得了结构化（“有类型”）策略，这类策略适合进行形式化验证，是迈向安全人工智能体的重要一步。此外，我们可以利用范畴逻辑，将智能体的目标描述为形式谓词，而这些谓词是否被满足可能取决于交互情境。这指向用于约束和控制自组织智能体集群的强大组合式工具。
英文摘要：
We introduce structured active inference, a large generalization and formalization of active inference using the tools of categorical systems theory. We cast generative models formally as systems "on an interface", with the latter being a compositional abstraction of the usual notion of Markov blanket; agents are then 'controllers' for their generative models, formally dual to them. This opens the active inference landscape to new horizons, such as: agents with structured interfaces (e.g. with 'mode-dependence', or that interact with computer APIs); agents that can manage other agents; and 'meta-agents', that use active inference to change their (internal or external) structure. With structured interfaces, we also gain structured ('typed') policies, which are amenable to formal verification, an important step towards safe artificial agents. Moreover, we can make use of categorical logic to describe express agents' goals as formal predicates, whose satisfaction may be dependent on the interaction context. This points towards powerful compositional tools to constrain and control self-organizing ensembles of agents.
Learning How to Remember: A Meta-Cognitive Management Method for Structured and Transferable Agent Memory
Institute of Automation, CAS（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Zhongguancun Academy（中关村学院）；Zhongguancun Institute of Artificial Intelligence（中关村人工智能研究所）；Department of Gastrointestinal Surgery, Fujian Provincial Cancer Hospital（福建省癌症医院胃肠外科）；College of Computer and Data Science, Fuzhou University（福州大学计算机与数据科学学院）
MedConclusion: A Benchmark for Biomedical Conclusion Generation from Structured Abstracts
Harvard AI and Robotics Lab, Harvard Medical School（哈佛大学医学院哈佛人工智能与机器人实验室）；University of Southern California（南加州大学）；Carnegie Mellon University（卡内基梅隆大学）；Stanford University（斯坦福大学）；Kempner Institute for the Study of Natural and Artificial Intelligence, Harvard University（哈佛大学肯普纳自然与人工智能研究所）
MathDoc: Benchmarking Structured Extraction and Active Refusal on Noisy Mathematics Exam Papers
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学人工智能学院）；Gaotu Techedu Inc.（高途科技公司）；Beijing Key Laboratory of Research on Large Models and Intelligent Governance（北京大模型与智能治理研究重点实验室）；Engineering Research Center of Next-Generation Intelligent Search and Recommendation, MOE（下一代智能搜索与推荐工程研究中心，教育部）；Nanjing University of Aeronautics and Astronautics（南京航空航天大学）；University of Science and Technology of China（中国科学技术大学）
PCA-Guided Autoencoding for Structured Dimensionality Reduction in Active Infrared Thermography
Khalifa University of Science and Technology（卡利法科技科学大学）；Advanced Research and Innovation Center（先进研究与创新中心）；Department of Aerospace Engineering（航空航天工程系）；BioMedIA Lab（BioMedIA实验室）；Mohamed Bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；Department of Electrical Engineering and Computer Science（电气工程与计算机科学系）；Department of Information Systems and Operations Research（信息系统与运营管理系）；Vienna University of Economics and Business（维也纳经济与商业大学）；Department of Industrial and Information Engineering and Economics (DIIIE)（工业与信息工程与经济学系）；University of L’Aquila（拉奎拉大学）；Department of Industrial and Systems Engineering（工业与系统工程系）
