<p>Measuring the influence of scientific articles has become a critical component of evaluating their value. Beyond existing indicators such as citation counts and novelty, the social media impact of scientific articles is gaining recognition as an important metric. However, the relationship between social media impact and existing indicators remains unclear. In particular, the correlation between social media impact and the novelty of scientific articles has not been thoroughly explored, and the correlation between social media impact and citation counts varies across research fields and post-publication stages. To address these gaps, we analyzed 5162 scientific articles mentioned by the Chinese Academy of Sciences on Weibo. Our study examined the correlation between social media impact and novelty across five research fields, as well as the correlation between social media impact and citation counts across five research fields and nine post-publication stages. We found no significant correlation between social media impact and novelty in any research field. Additionally, we observed a weak but significant correlation between social media impact and citation counts in natural sciences and engineering and technology, but no significant correlation in medical and health sciences, agricultural and veterinary sciences, or social sciences. Furthermore, social media impact was weakly correlated with citation counts across most post-publication stages, except at the nine-year stage. These findings provide valuable insights into the role of social media impact in evaluating the influence of scientific articles, highlighting its potential as a complementary metric in research evaluation.</p>