<p><strong>Duties:</strong></p><p>• Research and development of a pre-training framework that integrates medical text (electronic health records), images, sequential physiological signals, and other multimodal medical data</p><p>Frame.</p><p>• Explore cross-modal alignment techniques based on medical knowledge graphs, and develop complex medical</p><p>Learn reasoning and diagnostic techniques.</p><p>• Assist in scientific research project application work. </p><p>• Assist in the implementation and transformation of research results. </p><p>• Publish high-quality research papers (such as CVPR, ICCV, NeurIPS, ICLR, TPAMI, TMI, MedIA)</p><p>et al.), apply for a national patent for invention.</p><p></p><p><strong>Requirements:</strong></p><p></p><p>• Artificial intelligence, natural language processing, clinical medical informatics, and other related directions.</p><p>• Familiar with multimodal large models (such as CLIP, BLIP, LLaVA, InternVL, QwenVL, DeepSeek-R1)</p><p>etc.) development and application.</p><p>• Proficient in C/C++ and PyTorch/TensorFlow frameworks.</p><p>• Have strong research abilities, with top conference or journal paper publication experience.</p><p>• Those with medical scene research experience are preferred.</p>