Post
35
We reshaped how VLMs connect vision to language!
One simple architectural change ā dramatically better visual information injection into the LLM.
Accepted at NeurIPS. Check it out š
š https://arxiv.org/abs/2611.06844
š» https://github.com/YingyingYan1/DGP
#NeurIPS #VLM #MLLM #ViT
One simple architectural change ā dramatically better visual information injection into the LLM.
Accepted at NeurIPS. Check it out š
š https://arxiv.org/abs/2611.06844
š» https://github.com/YingyingYan1/DGP
#NeurIPS #VLM #MLLM #ViT
