Neuro-Symbolic Generative Diffusion Models for Physically Grounded, Robust, and Safe Generation
This addresses the problem of integrating generative AI into safety-critical or scientifically rigorous domains like drug discovery and materials engineering, representing a novel method rather than an incremental improvement.
The paper tackled the challenge of ensuring diffusion models comply with physical and operational constraints for safety-critical applications by introducing Neuro-Symbolic Diffusion (NSD), which interleaves diffusion with symbolic optimization to generate certifiably consistent samples for both continuous and discrete outputs, demonstrated in tasks like non-toxic molecular generation and collision-free trajectory optimization.
Despite the remarkable generative capabilities of diffusion models, their integration into safety-critical or scientifically rigorous applications remains hindered by the need to ensure compliance with stringent physical, structural, and operational constraints. To address this challenge, this paper introduces Neuro-Symbolic Diffusion (NSD), a novel framework that interleaves diffusion steps with symbolic optimization, enabling the generation of certifiably consistent samples under user-defined functional and logic constraints. This key feature is provided for both standard and discrete diffusion models, enabling, for the first time, the generation of both continuous (e.g., images and trajectories) and discrete (e.g., molecular structures and natural language) outputs that comply with constraints. This ability is demonstrated on tasks spanning three key challenges: (1) Safety, in the context of non-toxic molecular generation and collision-free trajectory optimization; (2) Data scarcity, in domains such as drug discovery and materials engineering; and (3) Out-of-domain generalization, where enforcing symbolic constraints allows adaptation beyond the training distribution.