现代情报2026，Vol.46Issue(5)：3-15,13.DOI:10.3969/j.issn.1008-0821.2026.05.001
生成式人工智能驱动哲学社会科学评价:理论逻辑与实践路径
Generative Artificial Intelligence-Driven Evaluation in Philosophy and Social Sciences:Theoretical Logic and Practical Pathways
摘要
Abstract
[Purpose/Significance]In recent years,research evaluation systems in the philosophy and social sciences(PSS)have faced increasing scrutiny worldwide due to their reliance on standardized,metric-based models that often fail to capture the epistemic,contextual,and value-laden nature of scholarly output.While digital technologies have trans-formed evaluation in STEM fields,their potential in philosophy and social sciences,particularly in Non-Western and policy-driven academic contexts,remains underexplored.In response to these challenges,this study introduces generative artificial intelligence(GAI)as a conceptual and methodological catalyst for rethinking the philosophy and social sciences evaluation system in China.It addresses a critical gap in current global scholarship:the lack of a theoretical framework that aligns AI-enabled evaluation tools with the specific knowledge structures and cultural logic of philosophy and social sciences disciplines.[Method/Process]This study adopted a theory-oriented approach grounded in conceptual analysis and policy review,rather than empirical testing.It first conducted a systematic literature review of national planning docu-ments,academic papers,and strategic evaluation policies to trace the developmental trajectory of philosophy and social sci-ences evaluation in China,identifying four key phases:qualitative assessment,metric-based quantification,multidimen-sional integration,and intelligent transformation.Building on this historical context,the study examined the affordances and limitations of generative AI in relation to the epistemological and normative characteristics of philosophy and social sci-ences evaluation.Representative GAI tools,such as CNKI's Youyan AI,Scite Assistant,and GPT-4,are referenced as analytical cases to illustrate how intelligent technologies reshape evaluation logic in concept and structure.Building on these insights,the study developed a three-dimensional model and outlines practical pathways for applying GAI in context-aware,value-sensitive evaluation systems for PSS in China.[Result/Conclusion]This study proposes a three-dimensional GAI-driven evaluation model characterized by intelligence,plurality,and dynamism.It argues that GAI's strengths in semantic understanding,knowledge processing,and structured reasoning provide a theoretical basis for supporting research assessment.The model highlights three mechanisms:intelligent evaluation improves interpretive depth and trans-parency;pluralistic value frameworks better reflect the normative orientation of PSS in China;and dynamic adaptation sup-ports process-based,real-time evaluation.These mechanisms are applicable to real-world tasks such as assessing policy relevance,tracking societal impact,and monitoring academic outputs across their lifecycle.The model thus offers both conceptual innovation and practical pathways for improving the quality,inclusiveness,and responsiveness of philosophy and social sciences evaluation.By aligning with national priorities like value sensitivity and category-based assessment,the framework contributes to the development of a research evaluation system with distinct Chinese characteristics.关键词
生成式人工智能/哲学社会科学评价/智能评价/人机协同/评价范式Key words
generative artificial intelligence/philosophy and social sciences evaluation/intelligent evaluation/human machine collaboration/evaluation paradigm分类
社会科学引用本文复制引用
张裕晨,吴翊昕,邱均平..生成式人工智能驱动哲学社会科学评价:理论逻辑与实践路径[J].现代情报,2026,46(5):3-15,13.基金项目
国家社会科学基金一般项目"大中小学一体化人工智能素养教育框架与图书馆培育机制研究"(项目编号:25BTQ019). （项目编号:25BTQ019）
