Current AGI research focuses heavily on scaling these foundation models and enhancing specific agent capabilities, such as complex reasoning and coding. However, despite this progress, even the most ...
Abstract: For long-horizon multi-task robotic manipulation, hierarchical approaches provide an effective way to combine high-level language-based task planning with low-level vision-language based sub ...
Abstract: In multi-task learning (MTL), gradient conflict poses a significant challenge. Effective methods for addressing this problem, including PCGrad, CAGrad, and GradNorm, in their original ...