在全球低碳转型与“双碳”目标深入推进的背景下，工业机器人作为智能制造的核心载体，其对碳排放效率的影响已成为学术界与政策制定者关注的焦点。本研究基于2007—2020年中国28个行业的面板数据，首先运用投入产出法测算各行业的隐含碳排放量，进而构建包含非期望产出的超效率SBM模型测算隐含碳排放效率，最后采用双向固定效应模型实证检验工业机器人渗透度对隐含碳排放效率的影响及其作用机制。研究结果表明：第一，样本期内中国各行业隐含碳排放效率均值为0.56，整体呈“U型”演变特征，行业间效率差距呈现先扩大后缩小的趋势；第二，工业机器人渗透度的提升显著促进了隐含碳排放效率的提高，机器人渗透度每提高1%，隐含碳排放效率平均提升5.6%。本研究为理解智能制造与低碳经济的协同机制提供了新的理论视角，也为政府制定差异化减排政策、企业优化技术投资决策提供了实证支撑。

Against the backdrop of the global low-carbon transformation and the deepening implementation of the "dual-carbon" goals, industrial robots, as the core carrier of intelligent manufacturing, have made their impact on carbon emission efficiency a focal point of attention for both academia and policymakers. Based on panel data from 28 industries in China spanning from 2007 to 2020, this study first employs the input-output method to measure the embodied carbon emissions of each industry. Subsequently, it constructs a super-efficiency SBM model incorporating undesirable outputs to assess embodied carbon emission efficiency. Finally, a two-way fixed effects model is utilized to empirically examine the impact of industrial robot penetration on embodied carbon emission efficiency and its underlying mechanisms. The research findings indicate that: First, during the sample period, the average embodied carbon emission efficiency across Chinese industries was 0.56, exhibiting an overall "U-shaped" evolutionary characteristic, with the efficiency gap between industries first widening and then narrowing. Second, an increase in industrial robot penetration significantly enhances embodied carbon emission efficiency; specifically, a 1% increase in robot penetration leads to an average improvement of 5.6% in embodied carbon emission efficiency. This study provides a new theoretical perspective for understanding the synergistic mechanism between intelligent manufacturing and the low-carbon economy, while also offering empirical support for governments to formulate differentiated emission reduction policies and for enterprises to optimize technology investment decisions.