Update README.md

README.md

@@ -6,7 +6,11 @@
 - **Chunked generation (best use case)**: Stable motion provides clean anchors AND makes the next chunk far more likely to correctly continue the direction of a given action
 - **Single generation**: Smooths motion reversal and "ping-pong" in 81+ frame generations
 
-See Demo Workflows in Folder and [demo clip here](https://youtu.be/wZgoklsVplc)
+Works with both **i2v** (image-to-video) and **t2v** (text-to-video), though i2v sees the most benefit due to anchor-based continuation.
+
+See Demo Workflows in Folder and demo clip below:
+
+[![Watch the demo](https://img.youtube.com/vi/wZgoklsVplc/maxresdefault.jpg)](https://youtu.be/wZgoklsVplc)
 
 If you find this useful:
 
@@ -22,10 +26,10 @@ Based on [FreeLong (NeurIPS 2024)](https://arxiv.org/abs/2407.19918), this imple
 
 | Without FreeLong | With FreeLong |
 |------------------|---------------|
-| Motion reverses mid-video ("ping-pong") | Stable directional motion throughout |
-| Subject drifts between "scenes" | Consistent subject appearance |
-| Prompt partially ignored for motion | Motion follows prompt intent |
-| 2-3 distinct "scenes" per generation | Single coherent sequence |
+| More prone to motion reversal ("ping-pong") | More likely to maintain stable direction |
+| More prone to subject drift between "scenes" | More likely to keep consistent appearance |
+| More prone to ignoring motion in prompt | More likely to follow prompt intent |
+| More prone to multiple distinct "scenes" | More likely to produce coherent sequence |
 
 **How it works:**
 - **Global stream**: Full-sequence attention captures overall motion direction