Downloads · 30 days
19
1% of all-time downloads
ViLAMP-llava-qwen
orange-sk/ViLAMP-llava-qwen
ViLAMP-llava-qwen is a machine learning model from orange-sk. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
ViLAMP is a video-language model for hour-long video understanding, addressing computational bottlenecks in long-form processing through differential distillation. It employs two mechanisms: (1) query-aware keyframe s…
All-time downloads
2.5K
Public
Parameters
8.2B
16.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors16.4 GB · 100%
At a glance
- License
- apache-2.0
- Model type
- llava_qwen
- Access
- Public
- Created
- Apr 1, 2025
- Updated
- Apr 4, 2025
- SHA
- b8969282
Model card
From the Hugging Face model README
ViLAMP-llava-qwen
ViLAMP is a video-language model for hour-long video understanding, addressing computational bottlenecks in long-form processing through differential distillation. It employs two mechanisms: (1) query-aware keyframe selection and (2) patch-level feature merging to preserve salient details in non-keyframes. ViLAMP achieves state-of-the-art performance on long-video benchmarks while enabling efficient processing of 10K-frame videos on a single GPU, balancing accuracy and computational efficiency.
- Type
- llava_qwen
- License
- apache-2.0
- Created
- Apr 1, 2025
- Updated
- Apr 4, 2025