Diffusion Twigs with Loop Guidance for Conditional Graph Generation
It addresses conditional graph generation for applications such as molecular design, though it appears incremental as an extension of diffusion methods.
The paper tackles conditional graph generation by introducing Twigs, a diffusion framework with multiple co-evolving flows and loop guidance, achieving strong performance gains over baselines in tasks like inverse molecular design.
We introduce a novel score-based diffusion framework named Twigs that incorporates multiple co-evolving flows for enriching conditional generation tasks. Specifically, a central or trunk diffusion process is associated with a primary variable (e.g., graph structure), and additional offshoot or stem processes are dedicated to dependent variables (e.g., graph properties or labels). A new strategy, which we call loop guidance, effectively orchestrates the flow of information between the trunk and the stem processes during sampling. This approach allows us to uncover intricate interactions and dependencies, and unlock new generative capabilities. We provide extensive experiments to demonstrate strong performance gains of the proposed method over contemporary baselines in the context of conditional graph generation, underscoring the potential of Twigs in challenging generative tasks such as inverse molecular design and molecular optimization.