Xia Fan

Corporate social responsibility (CSR) is shifting from compliance-driven to strategically embedded practice, yet large-scale identification of strategic CSR (SCSR) remains methodologically challenging due to reliance on surveys or ratings. This study proposes a hybrid identification framework combining pre-trained language models (PLMs) and Latent Dirichlet Allocation to conduct paragraph-level semantic analysis on 3022 CSR reports from Chinese manufacturing firms. The approach effectively identifies and extracts the topics of both strategic and non-strategic CSR. Findings show that only 25.6% of CSR behaviors simultaneously possess strategic intent, business synergy, resource investment, and value output. SCSR practices are highly concentrated on value chain optimization and R&D collaboration, with significant variations across ownership types and industry environments. This study contributes a novel PLM-based methodology for SCSR identification and offers empirical evidence and practical tools for CSR quality assessment, regulatory oversight, and strategic decision-making in manufacturing sectors.

The deep integration of sci-tech and industrial innovation, rooted in the fusion of exploratory and exploitative ambidextrous capabilities, is a common global challenge. Traditional actors like enterprises and universities struggle due to the inherent imbalance of ambidextrous capability. Developed countries (e.g., Germany’s Fraunhofer, Finland’s VTT) have achieved integration through new-type research organizations, but rely on a “static coordination” model across departments ill-suited for rapidly changing, multi-logic environments. In contrast, China’s new-type R&D institutions (NTRI), emerging as innovative organizations, are naturally equipped to handle such institutional complexity and have become key drivers of deep integration. This study takes NTRI as a longitudinal single-case study object. Based on ambidextrous innovation theory and resource action theory, it constructs an analytical framework of “identifying integration challenges—addressing integration challenges—achieving integrated innovation” to explore how NTRI build dynamic ambidextrous capability through resource actions to drive the internal mechanism of integrating sci-tech innovation and industrial innovation. The results show that: (1) Accurately identifying integration breakpoints, bottlenecks, and hurdles at different development phases and establishing integration goals are key prerequisites for achieving integrated innovation; (2) the process of achieving integrated innovation is essentially a dynamic reconstruction of ambidextrous capability, involving resource bricolage to reconfigure demand-driven ambidextrous linking capability, utilizing resource orchestration to fission context-synchronized ambidextrous integration capability, and executing resource concerto to leapfrog networked symbiotic ambidextrous empowerment capability; and (3) the integrated innovation of NTRI at different phases exhibits a dynamic evolution, evolving from unidirectional spillover-integrated innovation to bidirectional interactive integrated innovation, and ultimately to empowering symbiotic integrated innovation.