The stimulating impact of crop residue return on greenhouse gas (GHG) emissions from paddy fields have been widely accepted, while the influence of site environmental and human factors on the simulating degree remains unclear. Here, we performed a meta-analysis to assess the GHG emissions affected by residue return, and its mitigation potential combined with key factors in paddy fields. Drawing upon 1047 observation sets of CH4 and N2O emissions from 155 peer-reviewed publications we found that residue return to paddy fields caused an average increase of 73% CH4 emissions and 14% in N2O emissions. Utilizing meta-analytical models, we identified pH as the most significant driver modulating GHG emissions, followed by soil organic matter (SOC) and total nitrogen. In alkaline soils, combining straw return with intermittent irrigation (285.2%) or mid-season drainage (118.9%) significantly reduced CH4 emissions compared to continuous flooding (1201.9%). Additionally, pairing straw return with higher nitrogen inputs (above 150 kg N ha−1) improved soil N2O uptake by −11.5%. In acid and neutral soils, straw carbonization achieved soil CH4 negative emissions (from −2.9% to −39.3%), but the long-term effects remained unclear. Reduced drainage frequency mitigates N2O emissions but may increase CH4 emissions. To efficiently mitigate GHG emissions, we proposed low-carbon schemes for acid or neutral soils based on specific SOC content: For soils with SOC content <10 g kg−1, prioritize nitrogen input control with rates not exceeding 174 kg N ha−1. For soils with SOC content >10 g kg−1, prioritize adjusting the type of straw. Our study underscores the significance of site-specific factors in modulating GHG emissions. Efficient GHG mitigation can be achieved by combining residue return with other agronomic measures tailored to different soil conditions.