AI RESEARCH

Training-free Latent Inter-Frame Pruning with Attention Recovery

arXiv CS.CV

ArXi:2603.05811v1 Announce Type: new Current video generation models suffer from high computational latency, making real-time applications prohibitively costly. In this paper, we address this limitation by exploiting the temporal redundancy inherent in video latent patches. To this end, we propose the Latent Inter-frame Pruning with Attention Recovery (LIPAR) framework, which detects and skips recomputing duplicated latent patches. Additionally, we