# Scalable and Cost-Efficient de Novo Template-Based Molecular Generation

arXiv: https://arxiv.org/abs/2506.19865 · HTML: https://mlsanity.com/paper/2506.19865/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 54 references that are in the map.
- First posted on arXiv 10 Jun 2025.

## Builds on (most cited first, 30 of 31)

- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Equivariant Diffusion for Molecule Generation in 3D](https://mlsanity.com/paper/2203.17003/) (2022)
- [Unifying Count-Based Exploration and Intrinsic Motivation](https://mlsanity.com/paper/1606.01868/) (2016)
- [Graph Transformer Networks](https://mlsanity.com/paper/1911.06455/) (2019)
- [Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation](https://mlsanity.com/paper/2106.04399/) (2021)
- [#Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning](https://mlsanity.com/paper/1611.04717/) (2016)
- [GuacaMol: Benchmarking Models for De Novo Molecular Design](https://mlsanity.com/paper/1811.09621/) (2018)
- [Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development](https://mlsanity.com/paper/2102.09548/) (2021)
- [Count-Based Exploration with Neural Density Models](https://mlsanity.com/paper/1703.01310/) (2017)
- [Trajectory balance: Improved credit assignment in GFlowNets](https://mlsanity.com/paper/2201.13259/) (2022)
- [Optimization of Molecules via Deep Reinforcement Learning](https://mlsanity.com/paper/1810.08678/) (2018)
- [Biological Sequence Design with GFlowNets](https://mlsanity.com/paper/2203.04115/) (2022)
- [GFlowNets and variational inference](https://mlsanity.com/paper/2210.00580/) (2022)
- [Multi-Objective De Novo Drug Design with Conditional Graph Generative Model](https://mlsanity.com/paper/1801.07299/) (2018)
- [Multi-Objective GFlowNets](https://mlsanity.com/paper/2210.12765/) (2022)
- [Towards Understanding and Improving GFlowNet Training](https://mlsanity.com/paper/2305.07170/) (2023)
- [GFlowNets for AI-Driven Scientific Discovery](https://mlsanity.com/paper/2302.00615/) (2023)
- [Equivariant Flow Matching with Hybrid Probability Transport](https://mlsanity.com/paper/2312.07168/) (2023)
- [Top-Down Synthesis for Library Learning](https://mlsanity.com/paper/2211.16605/) (2022)
- [Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning](https://mlsanity.com/paper/2004.12485/) (2020)
- [Mixed Continuous and Categorical Flow Matching for 3D De Novo Molecule Generation](https://mlsanity.com/paper/2404.19739/) (2024)
- [Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design](https://mlsanity.com/paper/2110.06389/) (2021)
- [RGFN: Synthesizable Molecular Generation Using GFlowNets](https://mlsanity.com/paper/2406.08506/) (2024)
- [Molecular Design in Synthetically Accessible Chemical Space via Deep Reinforcement Learning](https://mlsanity.com/paper/2004.14308/) (2020)
- [Maximum entropy GFlowNets with soft Q-learning](https://mlsanity.com/paper/2312.14331/) (2023)
- [nach0: Multimodal Natural and Chemical Languages Foundation Model](https://mlsanity.com/paper/2311.12410/) (2023)
- [Generative Flows on Synthetic Pathway for Drug Design](https://mlsanity.com/paper/2410.04542/) (2024)
- [Pessimistic Backward Policy for GFlowNets](https://mlsanity.com/paper/2405.16012/) (2024)
- [SILVR: Guided Diffusion for Molecule Generation](https://mlsanity.com/paper/2304.10905/) (2023)
- [Generative Active Learning for the Search of Small-molecule Protein Binders](https://mlsanity.com/paper/2405.01616/) (2024)

## Built on by (most cited first, 5 of 5)

- [SynCoGen: Synthesizable 3D Molecule Generation via Joint Reaction and Coordinate Modeling](https://mlsanity.com/paper/2507.11818/) (2025)
- [Synthesizable Molecular Generation via Soft-constrained GFlowNets with Rich Chemical Priors](https://mlsanity.com/paper/2602.04119/) (2026)
- [Diffusion Models at the Drug Discovery Frontier: A Review on Generating Small Molecules versus Therapeutic Peptides](https://mlsanity.com/paper/2511.00209/) (2025)
- [CounterFlowNet: From Minimal Changes to Meaningful Counterfactual Explanations](https://mlsanity.com/paper/2602.17244/) (2026)
- [LipoAgent: Coordinating Fine-Tuned LLM Agents for Safer Lipid Design](https://mlsanity.com/paper/2605.25250/) (2026)

## Abstract (arXiv)

Template-based molecular generation offers a promising avenue for drug design by ensuring generated compounds are synthetically accessible through predefined reaction templates and building blocks. In this work, we tackle three core challenges in template-based GFlowNets: (1) minimizing synthesis cost, (2) scaling to large building block libraries, and (3) effectively utilizing small fragment sets. We propose Recursive Cost Guidance, a backward policy framework that employs auxiliary machine learning models to approximate synthesis cost and viability. This guidance steers generation toward low-cost synthesis pathways, significantly enhancing cost-efficiency, molecular diversity, and quality, especially when paired with an Exploitation Penalty that balances the trade-off between exploration and exploitation. To enhance performance in smaller building block libraries, we develop a Dynamic Library mechanism that reuses intermediate high-reward states to construct full synthesis trees. Our approach establishes state-of-the-art results in template-based molecular generation.