mlsanityData to 7 Oct 2026

Efficient and Programmable Exploration of Synthesizable Chemical Space

arXiv 2512.00384 ↗First posted 29 Nov 2025cs.LG q-bio.BMcited by 3

Not on the map. Why?

Abstract (arXiv)

The constrained nature of synthesizable chemical space poses a significant challenge for sampling molecules that are both synthetically accessible and possess desired properties. In this work, we present PrexSyn, an efficient and programmable model for molecular discovery within synthesizable chemical space. PrexSyn is based on a decoder-only transformer trained on a billion-scale datastream of synthesizable pathways paired with molecular properties, enabled by a real-time, high-throughput C++-based data generation engine. The large-scale training data allows PrexSyn to reconstruct the synthesizable chemical space nearly perfectly at a high inference speed and learn the association between properties and synthesizable molecules. Based on its learned property-pathway mappings, PrexSyn can generate synthesizable molecules that satisfy not only single-property conditions but also composite property queries joined by logical operators, thereby allowing users to ``program'' generation objectives. Moreover, by exploiting this property-based querying capability, PrexSyn can efficiently optimize molecules against black-box oracle functions via iterative query refinement, achieving higher sampling efficiency than even synthesis-agnostic baselines, making PrexSyn a powerful general-purpose molecular optimization tool. Overall, PrexSyn pushes the frontier of synthesizable molecular design by setting a new state of the art in synthesizable chemical space coverage, molecular sampling efficiency, and inference speed.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 26 of its 55 references that are in the map. First posted on arXiv 29 Nov 2025.

Builds on

26 of 55 references.

References · 26

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016Automatic chemical design using a data-driven continuous representation of moleculescited by 762
  3. 2021Flow Network based Generative Models for Non-Iterative Diverse Candidate Generationcited by 384
  4. 2018GuacaMol: Benchmarking Models for De Novo Molecular Designcited by 294
  5. 2013Learning Factored Representations in a Deep Mixture of Expertscited by 275
  6. 2017Molecular De Novo Design through Deep Reinforcement Learningcited by 264
  7. 2022Sample Efficiency Matters: A Benchmark for Practical Molecular Optimizationcited by 151
  8. 2019ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendationscited by 77
  9. 2020The Synthesizability of Molecules Proposed by Generative Modelscited by 67
  10. 2020Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learningcited by 57
  11. 2021Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Designcited by 44
  12. 2024RGFN: Synthesizable Molecular Generation Using GFlowNetscited by 39
14 more
  1. 2017Adaptive Sampled Softmax with Kernel Based Samplingcited by 38
  2. 2024Cut Your Losses in Large-Vocabulary Language Modelscited by 35
  3. 2020Molecular Design in Synthetically Accessible Chemical Space via Deep Reinforcement Learningcited by 32
  4. 2020Barking up the right tree: an approach to search over molecule synthesis DAGscited by 31
  5. 2023Genetic algorithms are strong baselines for molecule generationcited by 31
  6. 2024SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraintscited by 26
  7. 2024Generative Flows on Synthetic Pathway for Drug Designcited by 20
  8. 2024Projecting Molecules into Synthesizable Chemical Spacescited by 17
  9. 2024Saturn: Sample-efficient Generative Molecular Design using Memory Manipulationcited by 14
  10. 2025SynLlama: Generating Synthesizable Molecules and Their Analogs with Large Language Modelscited by 9
  11. 2025Exploring Synthesizable Chemical Space with Iterative Pathway Refinementscited by 7
  12. 2024Procedural Synthesis of Synthesizable Moleculescited by 6
  13. 2025SynCoGen: Synthesizable 3D Molecule Generation via Joint Reaction and Coordinate Modelingcited by 5
  14. 2024SynthFormer: Equivariant Pharmacophore-based Generation of Synthesizable Molecules for Ligand-Based Drug Designcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025SynCraft: Guiding Large Language Models to Predict Edit Sequences for Molecular Synthesizability Optimizationcited by 1
  2. 2026Project Ariadne: Prompt-Conditioned Route Generation for Synthesis Planningcited by 1
  3. 2026SynLaD: Latent Diffusion for Generating Synthesizable Molecules Conditioned on 3D Pharmacophore Profilescited by 0