SafetyGuard Collection Based on KT's Responsible AI principles, we are releasing a model on Hugging Face that mitigates harmful outputs generated by AI models. • 1 item • Updated Oct 2, 2025 • 10
See-Saw Modality Balance: See Gradient, and Sew Impaired Vision-Language Balance to Mitigate Dominant Modality Bias Paper • 2503.13834 • Published Mar 18, 2025 • 5
