single mention · model architecture
Video-frame pruning
Prunes video frames during inference; the cited implementation uses a pruning rate of 0.5.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See multimodal architecture for the methods filed beside it.
Also called Video-frame pruning during inference, video-pruning rate 0.5.
Filed under model architecture › multimodal architecture