HEGM-VK-LSVD-K12
short video watch time prediction
2 min read
A pretrained checkpoint of the HEGM (Hierarchical Exponential-Gaussian Mixtures) architecture for short video recommender systems. The model predicts the watch time distribution of a clip: the exponential component handles quick swipes, while a mixture of 12 Gaussian (normal) distributions handles engaged viewing, combining them via a trainable skip gate. The checkpoint was trained on the up0.01_ir0.01 subset of the VK-LSVD dataset and will be useful for researchers and developers of recommendation algorithms.
Key features
- Separates quick skips and engaged views using a trainable skip gate and 12 Gaussian components.
- Includes a ready-to-use PyTorch Lightning checkpoint (
last.ckpt) and the original experiment configconfig.yaml. - The series also includes variants with different numbers of components:
K=3,K=6, andK=9. - Custom PyTorch architecture: does not load directly via standard
transformers.AutoModel.from_pretrained.
What you can do
- Evaluate watch time prediction quality on the VK-LSVD dataset test split and reproduce paper results.
- Compare prediction accuracy across different mixture component counts (matching
K12againstK3,K6, andK9versions). - Use as a base model or baseline when developing ranking and recommendations for short video feeds.
Example
git clone [https://github.com/rw404/HEGM.git](https://github.com/rw404/HEGM.git)
cd HEGM
pip install -r requirements.txt
python scripts/download_artifacts.py --splits
python evaluate.py --config configs/vklsvd_k12.yaml --checkpoint checkpoints/vklsvd_k12/last.ckpt
Read more
DetailsMetrics on the VK-LSVD test set: MSE — 479.6837 sec², MAE — 15.7679 sec, XAUC — 0.6202.
Running evaluation requires processed data splits (including
meta.json) from the official repository.Dataset paths in
config.yaml are specified relative to the GitHub project root.









Comments