mlsanityData to 7 Oct 2026

Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization

arXiv 2206.12411 ↗First posted 22 Jun 2022cs.CE q-bio.BMcited by 151

Not on the map. Why?

Abstract (arXiv)

Molecular optimization is a fundamental goal in the chemical sciences and is of central interest to drug and material design. In recent years, significant progress has been made in solving challenging problems across various aspects of computational molecular optimizations, emphasizing high validity, diversity, and, most recently, synthesizability. Despite this progress, many papers report results on trivial or self-designed tasks, bringing additional challenges to directly assessing the performance of new methods. Moreover, the sample efficiency of the optimization--the number of molecules evaluated by the oracle--is rarely discussed, despite being an essential consideration for realistic discovery applications. To fill this gap, we have created an open-source benchmark for practical molecular optimization, PMO, to facilitate the transparent and reproducible evaluation of algorithmic advances in molecular optimization. This paper thoroughly investigates the performance of 25 molecular design algorithms on 23 tasks with a particular focus on sample efficiency. Our results show that most "state-of-the-art" methods fail to outperform their predecessors under a limited oracle budget allowing 10K queries and that no existing algorithm can efficiently solve certain molecular optimization problems in this setting. We analyze the influence of the optimization algorithm choices, molecular assembly strategies, and oracle landscapes on the optimization performance to inform future algorithm development and benchmarking. PMO provides a standardized experimental setup to comprehensively evaluate and compare new molecule optimization methods with existing ones. All code can be found at https://github.com/wenhao-gao/mol_opt.

Key facts. Cited by 151 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Jun 2022.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 11; 2023: 26; 2024: 36; 2025: 37; 2026: 41.

151 papers.

Sort:

By area:

  1. 2023Large Language Models on Graphs: A Comprehensive Surveycited by 190
  2. 2024Efficient Evolutionary Search Over Chemical Space with Large Language Modelscited by 59
  3. 2023Sample-efficient Multi-objective Molecular Optimization with GFlowNetscited by 48
  4. 2025GenMol: A Drug Discovery Generalist with Discrete Diffusioncited by 47
  5. 2022Reinforced Genetic Algorithm for Structure-based Drug Designcited by 46
  6. 2024Graph Diffusion Transformers for Multi-Conditional Molecular Generationcited by 44
  7. 2025El Agente: An Autonomous Agent for Quantum Chemistrycited by 43
  8. 2025Feedback Descent: Open-Ended Text Optimization via Pairwise Comparisoncited by 42
  9. 2024DrugAgent: Automating AI-aided Drug Discovery Programming through LLM Multi-Agent Collaborationcited by 40
  10. 2024Genetic-guided GFlowNets for Sample Efficient Molecular Optimizationcited by 36
  11. 2022Computer-Aided Multi-Objective Optimization in Small Molecule Discoverycited by 31
  12. 2023Genetic algorithms are strong baselines for molecule generationcited by 31
38 more
  1. 2024Multimodal Large Language Models for Inverse Molecular Design with Retrosynthetic Planningcited by 29
  2. 2025KL-Regularized Reinforcement Learning is Designed to Mode Collapsecited by 29
  3. 2024SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraintscited by 26
  4. 2024LICO: Large Language Models for In-Context Molecular Optimizationcited by 26
  5. 2025Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Modelscited by 23
  6. 2023ExPT: Synthetic Pretraining for Few-Shot Experimental Designcited by 22
  7. 2024A survey and benchmark of high-dimensional Bayesian optimization of discrete sequencescited by 22
  8. 2024Molecule Generation with Fragment Retrieval Augmentationcited by 22
  9. 2022Graph Neural Networks for Moleculescited by 21
  10. 2022Controllable Data Generation by Deep Learning: A Reviewcited by 20
  11. 2024Generative Flows on Synthetic Pathway for Drug Designcited by 20
  12. 2025LeMat-GenBench: A Unified Evaluation Framework for Crystal Generative Modelscited by 20
  13. 2023Drug Discovery with Dynamic Goal-aware Fragmentscited by 19
  14. 2025Latent Bayesian Optimization via Autoregressive Normalizing Flowscited by 19
  15. 2023Searching for High-Value Molecules Using Reinforcement Learning and Transformerscited by 18
  16. 2024Generative Artificial Intelligence for Navigating Synthesizable Chemical Spacecited by 18
  17. 2024Graph Reinforcement Learning for Combinatorial Optimization: A Survey and Unifying Perspectivecited by 16
  18. 2025GeLLMO: Generalizing Large Language Models for Multi-property Molecule Optimizationcited by 16
  19. 2023Multi-Fidelity Active Learning with GFlowNetscited by 15
  20. 2023Tanimoto Random Features for Scalable Molecular Machine Learningcited by 15
  21. 2023MatSciML: A Broad, Multi-Task Benchmark for Solid-State Materials Modelingcited by 14
  22. 2024ACEGEN: Reinforcement learning of generative chemical agents for drug discoverycited by 14
  23. 2024Saturn: Sample-efficient Generative Molecular Design using Memory Manipulationcited by 14
  24. 2024SmileyLlama: Modifying Large Language Models for Directed Chemical Space Explorationcited by 14
  25. 2023Conformal Drug Property Prediction with Density Estimation under Covariate Shiftcited by 13
  26. 2025Return of the Latent Space COWBOYS: Re-thinking the use of VAEs for Bayesian Optimisation of Structured Spacescited by 13
  27. 2023Lo-Hi: Practical ML Drug Discovery Benchmarkcited by 10
  28. 2025POLO: Preference-Guided Multi-Turn Reinforcement Learning for Lead Optimizationcited by 10
  29. 2024Diagnosing and fixing common problems in Bayesian optimization for molecule designcited by 9
  30. 2022Probabilistic Generative Transformer Language models for Generative Design of Moleculescited by 8
  31. 2022Re-evaluating sample efficiency in de novo molecule generationcited by 8
  32. 2023Molecule Design by Latent Prompt Transformercited by 8
  33. 2023Utilizing Reinforcement Learning for de novo Drug Designcited by 7
  34. 2023Augmented Memory: Capitalizing on Experience Replay to Accelerate De Novo Molecular Designcited by 7
  35. 2023GRAPES: Learning to Sample Graphs for Scalable Graph Neural Networkscited by 7
  36. 2025Exploring Synthesizable Chemical Space with Iterative Pathway Refinementscited by 7
  37. 2023Symmetric Replay Training: Enhancing Sample Efficiency in Deep Reinforcement Learning for Combinatorial Optimizationcited by 6
  38. 2023The power of motifs as inductive bias for learning molecular distributionscited by 6

50 most cited of 151 All 151 (JSON)