Transformers
Safetensors
fga_avsd_generation
audio-visual-scene-aware-dialog
video-question-answering
factor-graph-attention
multimodal
Instructions to use Idan/fga-avsd with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Idan/fga-avsd with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AVSDForResponseGeneration model = AVSDForResponseGeneration.from_pretrained("Idan/fga-avsd", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Seeded numbers; the grounding gap is resolved by projection width
Verified signature0ea23aa -
Predictions from the published checkpoint
Verified signature6fe6715 -
Widen the decoder projection to 1024: CIDEr 0.887, and the video ablation now bites
Verified signature37d4ab6 -
Model card
Verified signatured90e060 -
Generated answers for all 1,710 test questions
Verified signaturef739f1f -
AVSD answer generation, DSTC7
Verified signaturec0105de -
initial commit
Verified signature76f02e7