llm/transformers may actually be better than prior methods (not my area of expertise)

I did see a talk recently of fine-tuning a small LLM specifically for receipt transcription: https://www.youtube.com/watch?v=vDyxCHl-p8s

https://github.com/fw-ai/cookbook/tree/main/training/case-st...

HF has OCR LLM based models too, eg: https://huggingface.co/baidu/Unlimited-OCR