Vision-Language Models Create Cross-Modal Task Representations

Re-assign community