TAS Know: A Transparent, AI-powered Framework for Cross-disciplinary Knowledge Sharing

Background. As knowledge production becomes increasingly open and complex in the era of artificial intelligence (AI), ensuring explainability and interdisciplinary adaptability in knowledge sharing has emerged as a critical challenge. Objectives. This study proposes a transparent and explainable framework for cross-disciplinary knowledge sharing, called TAS Know, which addresses knowledge fragmentation, improves interpretability, and fosters trust in open knowledge environments. Methods. By integrating natural language processing, knowledge graph (KG) construction, explainable recommendation algorithm, and blockchain-based user contribution recording mechanism, the framework realizes the semantic organization, transparent dissemination and collaborative sharing of knowledge, and improves the efficiency and participation of users in acquiring knowledge. Discussion and Conclusion. TAS Know is intended to advance open science and educational resource sharing by providing an interpretable, trustable, and scalable mechanism for managing heterogeneous and multimodal knowledge resources. Its explainable recommendation component enhances user trust and understanding of system decisions, while the blockchain-driven incentive structure promotes accountability and sustained participation. Contributions. This work contributes to both theory and practice by addressing gaps in knowledge organization, promoting interdisciplinary collaboration, and advancing the development of next-generation knowledge services. Its practical implications include potential applications in digital libraries, educational platforms, and open research ecosystems, where scalability, transparency, and trustworthiness are essential.