Search papers, labs, and topics across Lattice.
Institute of Information Engineering, Chinese Academy of Sciences
2
0
4
Repurposing adversarial perturbations can significantly enhance continual learning, reducing forgetting and improving task transfer in large language models.
Token merging can be fast: MaMe accelerates ViTs by up to 2x and Stable Diffusion v2.1 by 31% using only GPU-friendly matrix operations, outperforming prior methods.
