Purpose <p>This study compared and ranked exercise interventions for improving cognitive function in adult patients with cancer.</p> Methods <p>Four databases were searched from inception to November 2025 for randomized controlled trials (RCTs) of exercise interventions in adult patients with cancer; the primary outcome was self-reported cognitive function. A frequentist network meta-analysis combined direct and indirect evidence, with interventions ranked using the surface under the cumulative ranking area. Study quality and evidence certainty were assessed using the Cochrane Risk of Bias Tool 2.0 and the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) methodology.</p> Results <p>Fifty-six RCTs involving 5297 participants and eight exercise interventions were included. Compared with controls, combined exercise, yoga, high-intensity interval training (HIIT), and walking showed significant benefits. Ranking identified walking and yoga as the most effective interventions, followed by HIIT. Global and local inconsistency tests showed no significant inconsistencies. Sensitivity analyses confirmed robust walking and combined exercise outcomes, whereas yoga and HIIT results were insufficiently stable. Overall evidence quality was low.</p> Conclusion <p>Walking was the most robust and effective intervention for improving self-reported cognitive function. Although yoga and HIIT showed benefits, their efficacy was not robust in sensitivity analyses and should be interpreted cautiously. These findings support prioritizing walking in clinical exercise prescriptions, whereas recommendations for yoga and HIIT require confirmatory evidence.</p> Implications for cancer survivors <p>Walking and yoga may improve self-reported cognitive function in cancer survivors. These low-cost, accessible exercises offer alternatives to pharmacological treatments, and personalized exercise plans may enhance recovery and quality of life.</p>