Zhihao Li

Stripping away seemingly helpful information from agents' observations can actually *improve* the robustness of multi-agent coordination in communication-constrained environments.

Ditch the quadratic cost and black-box nature of neural operators: Gaussian Particle Operators offer interpretable, near-linear PDE learning by representing fields as learned Gaussian atoms.

Finally, image generation can be controlled at the element level: adjust shapes, alter colors, or add/remove objects with this new vector graphics-guided framework.