mlsanityData to 14 Oct 2026

Guiding Generative Models to Uncover Diverse and Novel Crystals via Reinforcement Learning

arXiv 2511.07158 ↗First posted 10 Nov 2025cs.LG physics.comp-phcited by 21

On the map: Deep reinforcement learning methods → RL for energy, networks and operations

Abstract (arXiv)

Discovering functional crystalline materials entails navigating an immense combinatorial design space. While recent advances in generative artificial intelligence have enabled the sampling of chemically plausible compositions and structures, a fundamental challenge remains: the objective misalignment between likelihood-based sampling in generative modelling and targeted focus on underexplored regions where novel compounds reside. Here, we introduce a reinforcement learning framework that guides latent denoising diffusion models toward diverse and novel, yet thermodynamically viable crystalline compounds. Our approach integrates group relative policy optimisation with verifiable, multi-objective rewards that jointly balance creativity, stability, and diversity. Beyond de novo generation, we demonstrate enhanced property-guided design that preserves chemical validity, while targeting desired functional properties. This approach establishes a modular foundation for controllable AI-driven inverse design that addresses the novelty-validity trade-off across scientific discovery applications of generative models.

Key facts. Cited by 21 papers in the map (data through 14 Oct 2026). Placed in RL for energy, networks and operations, one of 15 directions in Deep reinforcement learning methods. Builds on 21 of its 63 references that are in the map. First posted on arXiv 10 Nov 2025.

Builds on

21 of 63 references.

References · 21

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2017Proximal Policy Optimization Algorithmscited by 15,215
  5. 2020Denoising Diffusion Implicit Modelscited by 9,781
  6. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  7. 2022Scalable Diffusion Models with Transformerscited by 6,054
  8. 2022Classifier-Free Diffusion Guidancecited by 5,847
  9. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
  10. 2020Improved Techniques for Training Score-Based Generative Modelscited by 1,043
  11. 2019Monte Carlo Gradient Estimation in Machine Learningcited by 283
  12. 2021Crystal Diffusion Variational Autoencoder for Periodic Material Generationcited by 270
9 more
  1. 2023Crystal Structure Prediction by Joint Equivariant Diffusioncited by 148
  2. 2024FlowMM: Generating Materials with Riemannian Flow Matchingcited by 109
  3. 2024Fine-Tuned Language Models Generate Stable Inorganic Materials as Textcited by 93
  4. 2024Large-scale Reinforcement Learning for Diffusion Modelscited by 78
  5. 2025All-atom Diffusion Transformers: Unified generative modelling of molecules and materialscited by 75
  6. 2025Reinforcement Learning from Human Feedbackcited by 58
  7. 2022Large-scale machine-learning-assisted exploration of the whole materials spacecited by 13
  8. 2025Reinforcement Fine-Tuning for Materials Designcited by 13
  9. 2025CrystalGym: A New Benchmark for Materials Discovery Using Reinforcement Learningcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 19.

21 papers.

Sort:

By area:

  1. 2026Open Materials Generation with Inference-Time Reinforcement Learningcited by 8
  2. 2026MADE: Benchmark Environments for Closed-Loop Materials Discoverycited by 7
  3. 2025Control Variate Score Matching for Diffusion Modelscited by 6
  4. 2026Zatom-1: Towards a Multimodal Foundation Model for 3D Molecules and Materialscited by 5
  5. 2025Generative Models for Crystalline Materialscited by 4
  6. 2026Crys-JEPA: Accelerating Crystal Discovery via Embedding Screening and Generative Refinementcited by 4
  7. 2026Substitution-Based Analysis of Structural Novelty for Generative Models of Materialscited by 3
  8. 2026Regret Analysis of Guided Diffusion for Black-Box Optimization over Structured Inputscited by 1
  9. 2026MatMind: A Structure-Activity Knowledge-Driven Generative Foundation Model for Materials Sciencecited by 1
  10. 2026Autoregressive latent diffusion for 3D molecule generationcited by 1
  11. 2026Chemical filters for ultra-high-throughput materials screening and generationcited by 1
  12. 2026CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Predictioncited by 1
9 more
  1. 2026Symmetry- and Property-Aware Crystal Generation with Reinforcement Learning for Inverse Materials Designcited by 1
  2. 2026OMatG-flash: An All-Atom Flow Map with Reinforce Adjoint Matching for Scalable Materials Discoverycited by 1
  3. 2026Representations from Pretrained Machine-Learning Interatomic Potentials as Coarse Coordinates for Material Generation and Evaluationcited by 0
  4. 2026On the Diversity of Analogy Making in Large Language Modelscited by 0
  5. 2026Let CSP Be Your ANCHOR: Adaptive Crystal Search over Frozen Structure Priorscited by 0
  6. 2026GLASS: Global Latent Aggregation with Slot-based Set Decoding for Scalable All-Atom Crystal Generationcited by 0
  7. 2026Riemannian Flow Models with Reinforcement Learning for Molecular Crystal Structure Predictioncited by 0
  8. 2026Reinforcement Learning on the Discrete Composition Channel of a Crystal Generator: Validated Gains and Reward Hackingcited by 0
  9. 2026DiffGA: Steering Frozen Diffusion Models for Materials Design via Evolutionary Searchcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Towards Automated Discovery: A Review of Generative Models, Multimodal Learning and Closed-Loop Workflows in Inverse Materials Design
  2. 2025Introducing physics-informed generative models for targeting structural novelty in the exploration of chemical space
  3. 2024Generative Inverse Design of Crystal Structures via Diffusion Models with Transformers
  4. 2025Establishing baselines for generative discovery of inorganic crystals
  5. 2026SVGym (SciVerseGym): An Environment for Reinforcement Learning and Bayesian Optimization in Crystal Discovery
  6. 2024Unleashing the power of novel conditional generative approaches for new materials discovery
  7. 2026CrystalReasoner: Reasoning and RL for Property-Conditioned Crystal Structure Generation
  8. 2025PLaID++: A Preference Aligned Language Model for Targeted Inorganic Materials Design