jiazhengli/deberta-large-asap_5

DeBERTa-Large ASAP Essay Scoring Model

This model is a fine-tuned version of microsoft/deberta-v3-large for automated essay scoring on the ASAP (Automated Student Assessment Prize) dataset, specifically trained on Question 5.

Model Description

• Model Type: Sequence Classification
• Architecture: DeBERTa-v3-Large (24 layers, 1024 hidden size, 16 attention heads)
• Task: Automated Essay Scoring
• Language: English
• Dataset: ASAP Question 5
• Number of Classes: 4 (score levels)

Dataset

This model was trained on the ASAP (Automated Student Assessment Prize) dataset, specifically Question 5. The ASAP dataset contains student essays with human-scored ratings.

• Dataset Source: ASAP-SAS Kaggle Competition
• Task: Score student essays on a predefined rubric
• Question 5 Prompt: The dataset contains essays written in response to specific prompts, with scores assigned by human raters

Training Details

• Base Model: microsoft/deberta-v3-large
• Training Data: ASAP dataset Question 5

License

Please refer to the original DeBERTa-v3-large model license and the ASAP dataset terms of use.