This repository contains the official training code for VFig, a vision-language model trained to generate SVG figures from scientific paper images. VFig/ ├── rl/ # GRPO-based reinforcement learning ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results