mlsanityData to 14 Oct 2026

Genetic-guided GFlowNets for Sample Efficient Molecular Optimization

arXiv 2402.05961 ↗First posted 5 Feb 2024q-bio.BM cs.LG cs.NEcited by 39

Not on the map. Why?

Abstract (arXiv)

The challenge of discovering new molecules with desired properties is crucial in domains like drug discovery and material design. Recent advances in deep learning-based generative methods have shown promise but face the issue of sample efficiency due to the computational expense of evaluating the reward function. This paper proposes a novel algorithm for sample-efficient molecular optimization by distilling a powerful genetic algorithm into deep generative policy using GFlowNets training, the off-policy method for amortized inference. This approach enables the deep generative policy to learn from domain knowledge, which has been explicitly integrated into the genetic algorithm. Our method achieves state-of-the-art performance in the official molecular optimization benchmark, significantly outperforming previous methods. It also demonstrates effectiveness in designing inhibitors against SARS-CoV-2 with substantially fewer reward calls.

Key facts. Cited by 39 papers in the map (data through 14 Oct 2026). Builds on 45 of its 67 references that are in the map. First posted on arXiv 5 Feb 2024.

Builds on

45 of 67 references.

References · 45

  1. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  2. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,367
  3. 2019Fine-Tuning Language Models from Human Preferencescited by 2,150
  4. 2016Automatic chemical design using a data-driven continuous representation of moleculescited by 765
  5. 2017Reinforcement Learning with Deep Energy-Based Policiescited by 669
  6. 2018Attention, Learn to Solve Routing Problems!cited by 656
  7. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 652
  8. 2016Neural Combinatorial Optimization with Reinforcement Learningcited by 608
  9. 2021Flow Network based Generative Models for Non-Iterative Diverse Candidate Generationcited by 392
  10. 2020POMO: Policy Optimization with Multiple Optima for Reinforcement Learningcited by 348
  11. 2018GuacaMol: Benchmarking Models for De Novo Molecular Designcited by 297
  12. 2021GFlowNet Foundationscited by 290
33 more
  1. 2017Molecular De Novo Design through Deep Reinforcement Learningcited by 267
  2. 2021Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Developmentcited by 264
  3. 2022Trajectory balance: Improved credit assignment in GFlowNetscited by 251
  4. 2018Optimization of Molecules via Deep Reinforcement Learningcited by 200
  5. 2017Bridging the Gap Between Value and Policy Based Reinforcement Learningcited by 182
  6. 2022Biological Sequence Design with GFlowNetscited by 174
  7. 2022Sample Efficiency Matters: A Benchmark for Practical Molecular Optimizationcited by 155
  8. 2022Learning GFlowNets from partial episodes for improved convergence and stabilitycited by 132
  9. 2020Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retrainingcited by 123
  10. 2022Sym-NCO: Leveraging Symmetricity for Neural Combinatorial Optimizationcited by 119
  11. 2022GFlowNets and variational inferencecited by 106
  12. 2018Multi-Objective De Novo Drug Design with Conditional Graph Generative Modelcited by 101
  13. 2022Exploring Chemical Space with Score-based Out-of-distribution Generationcited by 95
  14. 2021Learning Collaborative Policies to Solve NP-hard Routing Problemscited by 86
  15. 2023Better Training of GFlowNets with Local Credit and Incomplete Trajectoriescited by 85
  16. 2023Towards Understanding and Improving GFlowNet Trainingcited by 75
  17. 2019Augmenting Genetic Algorithms with Deep Neural Networks for Exploring the Chemical Spacecited by 66
  18. 2023GFlowNets for AI-Driven Scientific Discoverycited by 63
  19. 2020Guiding Deep Molecular Optimization with Genetic Explorationcited by 51
  20. 2023Sample-efficient Multi-objective Molecular Optimization with GFlowNetscited by 50
  21. 2023Local Search GFlowNetscited by 48
  22. 2022Reinforced Genetic Algorithm for Structure-based Drug Designcited by 46
  23. 2021Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Designcited by 44
  24. 2022A Variational Perspective on Generative Flow Networkscited by 39
  25. 2021JANUS: Parallel Tempered Genetic Algorithm Guided by Deep Neural Networks for Inverse Molecular Designcited by 32
  26. 2023Thompson sampling for improved exploration in GFlowNetscited by 32
  27. 2023Genetic algorithms are strong baselines for molecule generationcited by 31
  28. 2023Maximum entropy GFlowNets with soft Q-learningcited by 30
  29. 2023Learning to Scale Logits for Temperature-Conditional GFlowNetscited by 27
  30. 2023Learning Energy Decompositions for Partial Inference of GFlowNetscited by 24
  31. 2023Bootstrapped Training of Score-Conditioned Generator for Offline Design of Biological Sequencescited by 23
  32. 2023De novo Drug Design using Reinforcement Learning with Multiple GPT Agentscited by 19
  33. 2023Symmetric Replay Training: Enhancing Sample Efficiency in Deep Reinforcement Learning for Combinatorial Optimizationcited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 6; 2025: 11; 2026: 22.

39 papers.

Sort:

By area:

  1. 2024Efficient Evolutionary Search Over Chemical Space with Large Language Modelscited by 61
  2. 2024Flow of Reasoning: Training LLMs for Divergent Reasoning with Minimal Examplescited by 30
  3. 2024Adaptive teachers for amortized samplerscited by 23
  4. 2024Improved Off-policy Reinforcement Learning in Biological Sequence Designcited by 13
  5. 2024Beyond Squared Error: Exploring Loss Design for Enhanced Training of Generative Flow Networkscited by 11
  6. 2025ExLLM: Experience-Enhanced LLM Optimization for Molecular Design and Beyondcited by 8
  7. 2024Small Molecule Optimization with Large Language Modelscited by 6
  8. 2025Reinforced sequential Monte Carlo for amortised samplingcited by 6
  9. 2025Neural Genetic Search in Discrete Spacescited by 5
  10. 2025Refine Drugs, Don't Complete Them: Uniform-Source Discrete Flows for Fragment-Based Drug Discoverycited by 5
  11. 2026Graph-GRPO: Training Graph Flow Models with Reinforcement Learningcited by 5
  12. 2025Gradient GA: Gradient Genetic Algorithm for Drug Molecular Designcited by 4
27 more
  1. 2025A Genetic Algorithm for Navigating Synthesizable Molecular Spacescited by 4
  2. 2025Toward Closed-loop Molecular Discovery via Language Model, Property Alignment and Strategic Searchcited by 3
  3. 2025GP-MoLFormer-Sim: Test Time Molecular Optimization through Contextual Similarity Guidancecited by 2
  4. 2026Discrete diffusion samplers and bridges: Off-policy algorithms and applications in latent spacescited by 2
  5. 2026DrugR: Optimizing Molecular Drugs through LLM-based Explicit Reasoningcited by 2
  6. 2025MCCE: A Framework for Multi-LLM Collaborative Search in Discrete Spaces with Similarity-Filtered Preference Learningcited by 1
  7. 2025Iterative Foundation Model Fine-Tuning on Multiple Rewardscited by 1
  8. 2026SEISMO: Explanation-Aware, Trajectory-Conditioned LLM Agents for Sample-Efficient Molecular Optimisationcited by 1
  9. 2026Synthesizable Molecular Generation via Soft-constrained GFlowNets with Rich Chemical Priorscited by 1
  10. 2026Exploring Multiple High-Scoring Subspaces in Generative Flow Networkscited by 1
  11. 2026Stable GFlowNets with TV Monitoring and Probabilistic Guaranteescited by 1
  12. 2026Don't Retrain, Just Reuse: Recovering Dual-Target Molecules from Single-Target Diffusion Modelscited by 1
  13. 2026Sample Efficient Generative Optimization for Molecular Designcited by 1
  14. 2025Multi-Armed Sampling Problem and the End of Explorationcited by 0
  15. 2026Signal from Structure: Exploiting Submodular Upper Bounds in Generative Flow Networkscited by 0
  16. 2026From Tokens to Blocks: A Block-Diffusion Perspective on Molecular Generationcited by 0
  17. 2026INDIBATOR: Diverse and Fact-Grounded Individuality for Multi-Agent Debate in Molecular Discoverycited by 0
  18. 2026MolLIBRA: Genetic Molecular Optimization with Multi-Fingerprint Surrogates and Text-Molecule Aligned Criticcited by 0
  19. 2026Sample Efficient Generative Molecular Optimization with Joint Self-Improvementcited by 0
  20. 2026Amortized Molecular Optimization via Group Relative Policy Optimizationcited by 0
  21. 2026From Single-Step Edit Response to Multi-Step Molecular Optimizationcited by 0
  22. 2026FORGE: Fragment-Oriented Ranking and Generation for Context-Aware Molecular Optimizationcited by 0
  23. 2026Beyond Drug Discovery: The Nanotechnology Molecular Optimization (NMO) Benchmarkcited by 0
  24. 2026Oracle-Budgeted Molecular Optimization with Short-Term Graph Memorycited by 0
  25. 2026PMO-Dock: Benchmarking Docking, Specificity, and Generalization in Molecular Optimizationcited by 0
  26. 2026SNACK: A Sequential Notation Framework for Probabilistic Graph Generationcited by 0
  27. 2026MolHop: Scaffold-Hopping Molecular Design with Agentic Reinforcement Learningcited by 0