Last released Jul 7, 2026
PDF → OCR → sliding-window chunking → multi-question generation via LLM → JSONL dataset for fine-tuning
Supported by
