๐’๐จ๐ฅ๐ฎ๐ญ๐ข๐จ๐งไธจAI Text Detection with DeBERTa Fine-Tuning by Liu Chang๐’๐จ๐ฅ๐ฎ๐ญ๐ข๐จ๐งไธจAI Text Detection with DeBERTa Fine-Tuning by Liu Chang

๐’๐จ๐ฅ๐ฎ๐ญ๐ข๐จ๐งไธจAI Text Detection with DeBERTa Fine-Tuning

Liu Chang

Liu Chang

Lead AI/ML Engineer
๐’๐œ๐จ๐ฉ๐ž & ๐‚๐ก๐š๐ฅ๐ฅ๐ž๐ง๐ ๐ž: Built a DeBERTa-based classifier to classify human-written and AI-generated text. The key challenge was severe model overfitting during the fine-tuning phase.
๐ƒ๐ž๐ฅ๐ข๐ฏ๐ž๐ซ๐š๐›๐ฅ๐ž๐ฌ & ๐…๐ž๐ž๐๐›๐š๐œ๐ค: Architected a stabilized training pipeline with curriculum-based sampling and momentum-guided updates. Delivered a robust model achieving 92.5% accuracy, which now serves as a highly reliable baseline for the client's anti-plagiarism workflows.
Like this project

Posted Sep 15, 2026

Built a DeBERTa text classifier with curriculum-based sampling and momentum-guided updates, achieving 92.5% accuracy for anti-plagiarism workflows.

Likes

1

Views

0