Robustness Evaluation of OCR-based Visual Document Understanding under Multi-Modal Adversarial Attacks

Open in new window