Abstract: Building and improving collaborative innovation relationships between difference subjects is an important part of constructing the national innovation system. Under the impetus of new round of scientific and technological revolution, the boundaries between multiple innovation subjects are blurred. As a result, Industry-University-Research Collaborative Innovation (IURCI) has become increasingly important for the realization of S&T innovation, thus serves as a feasible way leading to the high-quality development of economy. Although the government itself cannot carry out innovation activities directly, it can play a supporting role in policy coordination, resource guarantee and other aspects to promote collaborative innovation. In the context of China’s economic transition which characterized by "big market - big government", government to provide investments and fundings for various innovation subjects is common phenomenon. As the essential measurements to promote collaborative innovation, Government Fundings (GF) have been proved to have positive effect on IURCI. However, less attention has been paid on how the GF works so that the exact functioning mechanism of GF hasn’t been revealed. S&T Social Organization (STSO) is an indispensable component of national innovation system. They can promote IURCI effectively by playing the role of "lubricant" and "glue" between the two forces of government and market. To reveal the impact and possible functioning mechanism of GF on IURCI, this paper focused on the direct effect of GF on IURCI as well as the mediation effect of S&T Social Organization. Based on the provincial panel data of 31 provinces in mainland China from 2010 to 2018, we calculated the collaborative innovation level of industry, university and research by using the collaborative degree model of compound collaborative system. Empirical study was also conducted using regression models, and the results has been tested with bootstrap method. The findings are as follows. First of all, the GF can effectively promote IURCI, but it doesn’t have direct effect on it. STSO serves as an intermediary mechanism and has fully transmitted the function of GF. Secondly, GF can promote the development of three types of STSO, but only the private non-enterprise units play the major role and completely mediate the impact of GF on IURCI. The findings above have revealed the mechanism of GF on IURCI by empirically testing the mediating effect of STSO during the process of government sponsoring the collaborative innovation. Those findings provide a possible explanation for the unstable empirical results of the role of government in previous studies. To meet the policy requirements in the 14th Five-year Plan, we further make the following two suggestions. First, establishing platform funding pattern to enhance the synergistic effect of GF. Second, strengthening the policy support for STSO to provide impetus for IURCI. The limitation of this paper mainly exists in the constraint of macro statistics. On the one hand, the lag of data makes it impossible to better represent the rich policy practice in recent years. On the other hand, the inaccessibility of data makes it difficult to subdivide the specific forms of GF. For future study, we can enrich the analysis of the mechanism of GF in promoting IURCI by exploring the heterogeneity of different types of government funding based on micro-data such as project funding.