Liucheng Guo

OPCNet not only enhances camouflaged object segmentation but also effectively filters out false positives in complex environments, redefining how we evaluate COD models.

Existing unlearning methods for MLLMs can lead to significant knowledge locality degradation, while the new SGPE method strikes a competitive balance between forgetting and preserving relevant knowledge.

AURORA-LM achieves state-of-the-art performance in language generation by leveraging a continuous-latent diffusion framework that preserves text fidelity while enhancing generative capabilities.