Downloads · 30 days
18
3% of all-time downloads
xlm-roberta-large-sinhala-multihead
akura-official/xlm-roberta-large-sinhala-multihead
xlm-roberta-large-sinhala-multihead is a machine learning model from akura-official. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
- Base model: XLM-RoBERTa Large - Task: Multi-head regression for Sinhala essay scoring - Dataset: 136 synthetic essays (ChatGPT-generated & scored) - Validation size: ~20 essays
All-time downloads
631
Public
Parameters
560M
9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin2.2 GB · 50%
At a glance
- Model type
- xlm-roberta
- Access
- Public
- Created
- Dec 10, 2025
- Updated
- Jan 20, 2026
- SHA
- 5d50c307
Model card
From the Hugging Face model README
Training Notes (Phase 5)
- Base model: XLM-RoBERTa Large
- Task: Multi-head regression for Sinhala essay scoring
- Dataset: 136 synthetic essays (ChatGPT-generated & scored)
- Validation size: ~20 essays
Results
- Training loss converged: 1.17 → 0.42
- MAE: 1.00
- RMSE: 1.23
- Pearson r: 0.58
Reminder: This model is a feasibility prototype. Future work includes retraining on real teacher-scored essays.
- Type
- xlm-roberta
- Created
- Dec 10, 2025
- Updated
- Jan 20, 2026