FastSTAR: Spatiotemporal Token Pruning for Efficient Autoregressive Video Synthesis
Open paper
Benchmark
Model
Rank
Results
text-to-video-generation-on-vbench
FastSTAR (20%, 30%, 40%, 70%)
–
VBench Total Score: 83.18
text-to-video-generation-on-vbench
FastSTAR (20%, 30%, 50%, 60%)
–
VBench Quality Score: 83.71
Rank counts only results with a code link.