Specialized Small Language Models; Optimized Fine-Tuning; Efficient Inference
Extract text and tables from document images
