The COVID-19 pandemic has substantially affected college students’ mental health worldwide. Although daily life has gradually returned to normal, less is known about whether the underlying structure of psychological symptoms—and their associations with personality traits—has shifted in the post-pandemic period. Addressing this gap, the present study applied network analysis to psychological and personality data from 51,305 college freshmen to compare pre-pandemic and post-pandemic cohorts. Using the Eysenck Personality Questionnaire (EPQ) and the Symptom Checklist-90 (SCL-90), networks based on EPQ dimensions and SCL-90 subscale scores were estimated and compared across pre-pandemic and post-pandemic cohorts. Results indicated that anxiety and depression remained central symptom domains across both periods; however, the network-structure invariance test was significant. Specifically, the expected influence (EI) of both interpersonal sensitivity (ΔEI = 0.058, p = 0.007) and paranoia (ΔEI = 0.059, p = 0.007) showed a significant increase. The regularized partial associations between symptoms and personality traits also changed, with the negative correlation between extraversion and interpersonal sensitivity weakening, and the positive correlation between neuroticism and obsessive–compulsive symptoms strengthening. These findings suggest that although overall mental health showed no substantial deterioration compared to the pre-pandemic period, the underlying symptom network and its associations with personality traits differ between cohorts, highlighting the potential value of considering personality-related information in future research on post-pandemic mental health among university students.