Return to Article Details Bridging the gap between machine translation output and images in multimodal documents Download Download PDF