Add links to paper and project page
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,23 +1,27 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
license: other
|
| 3 |
license_name: base-model-terms
|
| 4 |
pipeline_tag: image-to-video
|
| 5 |
tags:
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
base_model:
|
| 12 |
-
- nvidia/Lyra-2.0
|
| 13 |
-
- robbyant/lingbot-world-v2-14b-causal-fast
|
| 14 |
-
- Drexubery/UniView
|
| 15 |
-
datasets:
|
| 16 |
-
- ziqima/TrajectoryBench
|
| 17 |
---
|
| 18 |
|
| 19 |
# LoGo checkpoints
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
The LoRA adapter each model was post-trained to with the LoGo reward.
|
| 22 |
|
| 23 |
| file | model | base weights |
|
|
@@ -41,4 +45,4 @@ Usage, with the code at https://github.com/ziqi-ma/logo.
|
|
| 41 |
--rl-ckpt logo-uniworld-view-lora.safetensors --pose-scale 1.5
|
| 42 |
|
| 43 |
Evaluation on TrajectoryBench, including the per-group generation settings, is described in
|
| 44 |
-
`eval/README.md` in that repository.
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model:
|
| 3 |
+
- nvidia/Lyra-2.0
|
| 4 |
+
- robbyant/lingbot-world-v2-14b-causal-fast
|
| 5 |
+
- Drexubery/UniView
|
| 6 |
+
datasets:
|
| 7 |
+
- ziqima/TrajectoryBench
|
| 8 |
license: other
|
| 9 |
license_name: base-model-terms
|
| 10 |
pipeline_tag: image-to-video
|
| 11 |
tags:
|
| 12 |
+
- world-model
|
| 13 |
+
- video-generation
|
| 14 |
+
- camera-control
|
| 15 |
+
- reinforcement-learning
|
| 16 |
+
- lora
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
---
|
| 18 |
|
| 19 |
# LoGo checkpoints
|
| 20 |
|
| 21 |
+
**Paper:** [LoGo: Local-Global Rewards for Consistent Long-Horizon Video Generation](https://huggingface.co/papers/2610.03636)
|
| 22 |
+
|
| 23 |
+
**Project page:** [ziqi-ma.github.io/logo-website](https://ziqi-ma.github.io/logo-website)
|
| 24 |
+
|
| 25 |
The LoRA adapter each model was post-trained to with the LoGo reward.
|
| 26 |
|
| 27 |
| file | model | base weights |
|
|
|
|
| 45 |
--rl-ckpt logo-uniworld-view-lora.safetensors --pose-scale 1.5
|
| 46 |
|
| 47 |
Evaluation on TrajectoryBench, including the per-group generation settings, is described in
|
| 48 |
+
`eval/README.md` in that repository.
|