Update docs/source/en/api/pipelines/cogvideox.md

Co-authored-by: Steven Liu <[email protected]>

Author: a-r-r-o-w

docs/source/en/api/pipelines/cogvideox.md

@@ -104,7 +104,7 @@ It is also worth noting that torchao quantization is fully compatible with [torc
 
 [Pyramid Attention Broadcast](https://huggingface.co/papers/2408.12588) from Xuanlei Zhao, Xiaolong Jin, Kai Wang, Yang You.
 
-Pyramid Attention Broadcast (PAB) is a method that speeds up inference in diffusion models by systematically skipping attention computations between successive inference steps, and re-using cached attention states. This is due to the realization that the attention states do not differ too much numerically between successive steps. This difference is most significant/prominent in the spatial attention blocks, lesser so in temporal attention blocks, and least in cross attention blocks. Therefore, many cross attention computation blocks can be skipped, followed by temporal and spatial attention blocks. By combining other techniques like Sequence Parallelism and CFG Parallelism, the authors achieve near real-time video generation.
+Pyramid Attention Broadcast (PAB) is a method that speeds up inference in diffusion models by systematically skipping attention computations between successive inference steps and reusing cached attention states. The attention states aren't that different between successive steps. The most prominent difference is in the spatial attention blocks, not as much in the temporal attention blocks, and finally the least in the cross attention blocks. Therefore, many cross attention computation blocks can be skipped, followed by the temporal and spatial attention blocks. By combining other techniques like sequence parallelism and classifier-free guidance parallelism, PAB achieves near real-time video generation.
 
 PAB can be enabled easily on any pipeline by deriving from the [`PyramidAttentionBroadcastMixin`] and keeping track of current inference timestep in the pipeline. Minimal example to demonstrate how to use PAB with CogVideoX: