7,361 (AI text → Student text) pairs for training an AI text humanizer.
Each sample is a chat-formatted conversation where the model learns to rewrite formal AI-generated text into natural student voice.
Source: PERSUADE 2.0 Corpus — 25,996 pre-ChatGPT student essays
Filter: Holistic essay score 4-6, word count 100-600 → 7,361 essays
AI Rewrite: Each student essay was rewritten in… See the full description on the dataset page:
https://huggingface.co/datasets/changcheng967/Aegis-SFT-v3-Dataset.