FragFlex: Fragment-Based Flexible Molecular Generation for Exploring Vast Chemical Space

Sihyun Park; David Ryan Koes

cs.LGOct 11, 2026

FragFlex: Fragment-Based Flexible Molecular Generation for Exploring Vast Chemical Space

Authors: Sihyun Park, David Ryan Koes

Organizations: Carnegie Mellon University · University of Pittsburgh

Abstract

Drug discovery requires efficient exploration of an astronomically large chemical space. Molecular graph diffusion models offer a promising approach to this task, but many require graph size to be fixed before generation. Although atom-level insertion and deletion can enable flexible-size generation, jointly modeling molecular size, atom types, and bonds while maintaining chemical validity remains challenging. We introduce FragFlex, a flexible-length discrete diffusion model that performs node insertion and deletion at the fragment level. During denoising, FragFlex jointly learns the size, fragment identities, and inter-fragment connectivity of the fragment graph. A neural attachment-site predictor and valence-aware assembler then convert the generated fragment graph into an atom-level molecule. In large-scale experiments generating 10 million molecules, FragFlex achieves near-perfect chemical validity while maintaining high uniqueness, demonstrating a strong balance between the two.

Figures & tables

Figure 1: Overview of FragFlex. An atom-level molecule $M$ is preprocessed into a clean fragment graph $G_{0}$ . Forward diffusion applies categorical noise $\bar{Q}_{t}$ and structural deletion $\zeta^{\prime}(t)$ while protecting one root, yielding the single-fragment state $G_{T}$ . During sampling, $g_{\phi}$ predicts the insertion count $\widehat{\Delta}_{t}$ , and $f_{\theta}$ predicts the node types and edge connections of the added $\textsc{del}^{*}$ placeholders. At the final denoising step, predicted attachment sites guide the valence-aware assembly of the final molecule.

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Symbol	Meaning
$M$	Atom-level molecule.
$\mathcal{V}=\{F_{1},\ldots,F_{K}\}$	Fixed vocabulary of $K$ molecular fragments.
$G_{0}=(X_{0},E_{0},S_{0})$	Clean fragment graph with fragment labels, binary fragment edges, and directed attachment sites.
$G_{t}$	Fragment graph after structural deletion and categorical corruption at timestep $t$ .
$V_{0}=\{1,\ldots,N\}$	Node set of the clean fragment graph.
$A_{i}$	Number of atoms in the fragment assigned to node $i$ .

Appendix

Table 2: Notation used in the main text and Appendix A .

Component	Setting
Diffusion	$T=500$ with a cosine noise schedule $\bar{\alpha}_{t}$ ( $s=0.008$ )
Deletion/insertion schedule	Logistic $\zeta^{\prime}(t)$ with $s_{\mathrm{p}}=0$ , $w=0.05$
Terminal prior $G_{T}$	One fragment from the empirical fragment marginal
Main denoiser $f_{\theta}$	7 layers, node width 384, edge width 96, 8 heads
Count model $g_{\phi}$	2 layers, node width 128, edge width 32, 8 heads
Maximum graph size	$N_{\max}=32$ fragments

Appendix

Table 5: Hyperparameters and model configuration

Figure 4: The 100 fragment types used by FragFlex, arranged in descending order of frequency in the ZINC training set.

Figure 5: 50 random samples generated by FragFlex.

Explore similar work

Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation
Sep 11, 2026Thao Nguyen, Jeonghwan Kim, Zhenhailong Wang +1Molecular Generation Molecular Graph Generation
Siebel School of Computing and Data Science University of Illinois Urbana-Champaign Urbana, IL 61801, USA
MolHIT: Advancing Molecular-Graph Generation with Hierarchical Discrete Diffusion Models
Feb 19, 2026Hojung Jung, Rodrigo Hormazabal, Jaehyeong Jo +5Molecular Graph Generation Diffusion Models
KAIST AI · LG AI Research · Seoul National University
FRIGID: Scaling Diffusion-Based Molecular Generation from Mass Spectra at Training and Inference Time
Apr 17, 2026Montgomery Bohde, Hongxuan Liu, Mrunali Manjrekar +4Inference-Time Scaling Molecular Generation
Massachusetts Institute of Technology, Cambridge, MA, United States · Texas A&M University, College Station, TX, United States

Sep 11, 2026J/K move · Enter open · S save

Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation

Thao Nguyen, Jeonghwan Kim, Zhenhailong Wang, Heng Ji

Siebel School of Computing and Data Science University of Illinois Urbana-Champaign Urbana, IL 61801, USA

Molecular Generation Molecular Graph Generation Autoregressive Generation Molecular Optimization Conditional Generative Modeling

Open paper AskPDF arXiv

We introduce Fraglingo, an autoregressive molecular generator that constructs molecules step by step from chemically meaningful fragments connected through predefined attachment sites. At each generation step, Fraglingo jointly predicts which fragment to add and how it should attach by producing an attachment-aware fragment embedding and retrieving the nearest fragment through latent-space search. A wildcard-anchored readout represents both the growing molecule and candidate fragments relative to their attachment sites, enabling a single latent prediction to determine both fragment identity and attachment configuration. Because prediction operates in a continuous embedding space rather than over fixed fragment identifiers, larger fragment libraries can be introduced at inference time without retraining. This retrieval-based formulation provides a unified generation primitive for molecule generation, scaffold generation, scaffold decoration, and molecule optimization. Fraglingo also supports property-conditional generation, allowing desired molecular properties to guide the generation process. On controlled property-conditional benchmarks, Fraglingo achieves stronger joint property control than comparably trained baselines while maintaining competitive validity, uniqueness, and novelty. It also generalizes to fragment libraries up to four times larger than those used during training without retraining. Code is available at: https://anonymous.4open.science/r/FragLingo-3551.

Abstract

Explore similar work

Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation

MolHIT: Advancing Molecular-Graph Generation with Hierarchical Discrete Diffusion Models

FRIGID: Scaling Diffusion-Based Molecular Generation from Mass Spectra at Training and Inference Time