This monograph proposes a radical reframing of human history through the lens of coordination—the mechanisms by which billions of strangers align their actions. Arguing that coordination, rather than energy, capital, or the state, is the primary technology of civilization, the book traces the evolution of this technology from collective memory in small communities to money as a universal "information compressor." The author demonstrates that money, while a brilliant engineering compromise that enabled global scale, achieved this by radically simplifying reality. In the monetary signal, human talent, integrity, long-term consequences, and pedagogical contributions disappear. Society pays for this compression with an "economy of excess costs"—a gigantic layer of intermediaries (financial, legal, bureaucratic) that service not production, but the limitations of the information transmission mechanism itself. Building on the historical socialist calculation debate (Mises, Hayek) and recent advances in agent-based modeling, the book posits that artificial intelligence offers a way beyond this historical compromise. AI is framed not as a digital dictator or a replacement for human labor, but as a potential new coordination technology capable of multidimensional accounting and direct feedback. The author outlines a "two-circuit architecture" for the future, separating strategic political competition from everyday life, ensuring systemic stability even under self-interested elites. This work is intended for scholars and readers in economic history, philosophy of technology, institutional economics, and science and technology studies (STS) seeking to understand the past and future of civilization beyond conventional debates about money and power.
GAOKAO-Bench is introduced, an intuitive benchmark that employs questions from the Chinese GAOKAO examination as test samples, including both subjective and objective questions that contribute a robust evaluation benchmark for future large language models and offers valuable insights into the advantages and limitations of such models.
Xiaotian Zhang, Chun-yan Li, Yi Zong et al.· arXiv.org· 216 citations· ⚡17
Empirically, PRISM reduces the end-to-end time for data selection and model tuning to just 30% of conventional pipelines, and achieves this efficiency while simultaneously enhancing performance, surpassing models fine-tuned on the full dataset across eight multimodal and three language understanding benchmarks.
Jinhe Bi, Yifan Wang, Danqi Yan et al.· arXiv.org· 73 citations· ⚡4
The method, ECCOLA, is presented, which aims at making the high-level AI ethics principles more practical, making it possible for developers to more easily implement them in practice.
Ville Vakkuri, Kai-Kristian Kemell, P. Abrahamsson· EUROMICRO Conference on Soft...· 64 citations· ⚡6
This paper designs Markov decision processes (MDPs) for different combinatorial problems and proposes to train conditional GFlowNets to sample from the solution space and demonstrates that GFlowNet policies can efficiently find high-quality solutions.
Dinghuai Zhang, H. Dai, Esmeralda S. Whitammer et al.· Advances in Neural Informati...· 59 citations· ⚡8
An empirical study on the current state of practice in artificial intelligence ethics is conducted by means of a multiple case study of five case companies, which indicates a gap between research and practice in the area.
Ville Vakkuri, Kai-Kristian Kemell, Joni Kultanen et al.· arXiv.org· 56 citations· ⚡6