aiqbench analysis
This paper introduces Neuro-Symbolic Graph Generative Modeling (NSGGM), a framework that combines autoregressive neural proposals with SMT solvers to enforce hard logical constraints in graph generation tasks. By decoupling structural proposal from discrete assembly, the method achieves high satisfaction rates on benchmarks like MolSAT where purely neural approaches fail under out-of-distribution conditions. This advancement is critical for high-stakes applications such as drug discovery, where formal guarantees are necessary rather than probabilistic approximations.
Neural Proposals, Symbolic Guarantees: Neuro-Symbolic Graph Generative Modeling
arXiv:2602.16954v3 Announce Type: replace Abstract: While deep generative models excel at capturing graph data distributions, they struggle to satisfy complex, hard constraints. In unconstrained settings, these models typically produce valid topologies; yet imposing strict compositional rules, like those in drug discovery, creates an out-of-distribution (OOD) setting where purely neural methods frequently fail. Because these neural approaches rely on soft conditioning and post-hoc filtering on such tasks, they cannot provide the formal guarantees needed for high-stakes domains. To address this, we introduce Neuro-Symbolic Graph Generative Modeling (NSGGM), a framework built on the principle of Neural Proposals, Symbolic Guarantees. NSGGM decouples generation: an autoregressive model proposes structural scaffolds, and a Satisfiability Modulo Theories (SMT) solver handles the final discrete assembly of the proposed substructures. Empirically, NSGGM is competitive with state-of-the-art methods on unconstrained tasks. To evaluate logical-constraint satisfaction inspired by real drug discovery workflows, we introduce MolSAT, a benchmark for hard compositional rules. On MolSAT, purely neural baselines completely fail OOD (0% satisfaction with zero training support), while NSGGM achieves >95% satisfaction in-distribution and 64-86% with zero training support.
Abstract:While deep generative models excel at capturing graph data distributions, they struggle to satisfy complex, hard constraints. In unconstrained settings, these models typically produce valid topologies; yet imposing strict compositional rules, like those in drug discovery, creates an out-of-distribution (OOD) setting where purely neural methods frequently fail. Because these neural approaches rely on soft conditioning and post-hoc filtering on such tasks, they cannot provide the formal guarantees needed for high-stakes domains. To address this, we introduce Neuro-Symbolic Graph Generative Modeling (NSGGM), a framework built on the principle of Neural Proposals, Symbolic Guarantees. NSGGM decouples generation: an autoregressive model proposes structural scaffolds, and a Satisfiability Modulo Theories (SMT) solver handles the final discrete assembly of the proposed substructures. Empirically, NSGGM is competitive with state-of-the-art methods on unconstrained tasks. To evaluate logical-constraint satisfaction inspired by real drug discovery workflows, we introduce MolSAT, a benchmark for hard compositional rules. On MolSAT, purely neural baselines completely fail OOD (0% satisfaction with zero training support), while NSGGM achieves >95% satisfaction in-distribution and 64-86% with zero training support.
Machine Learning (cs.LG)
arXiv:2602.16954 [cs.LG]
(or arXiv:2602.16954v3 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2602.16954 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 264 words
Related coverage
A Neuro-Symbolic Approach for Probabilistic Reasoning on Graph Data
arXiv · Jul 20, 2026
ANNEAL: Adapting LLM Agents via Governed Symbolic Patch Learning
arXiv · Jun 9, 2026
Constraint-Anchored Reasoning Traces
arXiv · Jul 21, 2026
Constraint-Guided Enterprise Data Mapping with Large Language Models
arXiv · Aug 27, 2026