Abstract: Large language models (LLMs) have significantly enhanced cross-modal understanding capabilities by integrating visual encoders with textual embeddings, giving rise to multimodal large ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results