caiyijun

use conv layer to perform regression instead of fully-connected layer?

1

The original YOLO use two fylly-connected layers to perform regression, while this implementation use conv layer instead. What's the point of this change? Can conv regression capture global information as...

issue on activating thinking mode with lmdeploy

2

Hi, thanks for sharing the InternVL3.5 series! The thinking mode can be activated by setting the system prompt when inferecing with transformers, but how should it be done when running...