# FRAGMENTA: End-to-end Fragmentation-based Generative Model with Agentic Tuning for Drug Lead Optimization

arXiv: https://arxiv.org/abs/2511.20510 · HTML: https://mlsanity.com/paper/2511.20510/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical and scientific domain agents, one of 16 directions in LLM agents and tool use.
- Builds on 26 of its 49 references that are in the map.
- First posted on arXiv 25 Nov 2025.

## Placed in

- LLM agents and tool use / [Medical and scientific domain agents](https://mlsanity.com/agents/medical-and-scientific-domain-agents/)

## Builds on (most cited first, 26 of 26)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [A Point Set Generation Network for 3D Object Reconstruction from a Single Image](https://mlsanity.com/paper/1612.00603/) (2016)
- [Junction Tree Variational Autoencoder for Molecular Graph Generation](https://mlsanity.com/paper/1802.04364/) (2018)
- [GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation](https://mlsanity.com/paper/2203.02923/) (2022)
- [GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders](https://mlsanity.com/paper/1802.03480/) (2018)
- [GuacaMol: Benchmarking Models for De Novo Molecular Design](https://mlsanity.com/paper/1811.09621/) (2018)
- [Molecular De Novo Design through Deep Reinforcement Learning](https://mlsanity.com/paper/1704.07555/) (2017)
- [Hierarchical Generation of Molecular Graphs using Structural Motifs](https://mlsanity.com/paper/2002.03230/) (2020)
- [GraphNVP: An Invertible Flow Model for Generating Molecular Graphs](https://mlsanity.com/paper/1905.11600/) (2019)
- [Diffusion-based Molecule Generation with Informative Prior Bridges](https://mlsanity.com/paper/2209.00865/) (2022)
- [Generating 3D Molecules for Target Protein Binding](https://mlsanity.com/paper/2204.09410/) (2022)
- [Retro*: Learning Retrosynthetic Planning with Neural Guided A* Search](https://mlsanity.com/paper/2006.15820/) (2020)
- [Learning to Extend Molecular Scaffolds with Structural Motifs](https://mlsanity.com/paper/2103.03864/) (2021)
- [DrugAssist: A Large Language Model for Molecule Optimization](https://mlsanity.com/paper/2401.10334/) (2023)
- [Symmetry-Aware Actor-Critic for 3D Molecular Design](https://mlsanity.com/paper/2011.12747/) (2020)
- [De Novo Molecular Generation via Connection-aware Motif Mining](https://mlsanity.com/paper/2302.01129/) (2023)
- [ChemTS: An Efficient Python Library for de novo Molecular Generation](https://mlsanity.com/paper/1710.00616/) (2017)
- [Barking up the right tree: an approach to search over molecule synthesis DAGs](https://mlsanity.com/paper/2012.11522/) (2020)
- [Rigid Body Flows for Sampling Molecular Crystal Structures](https://mlsanity.com/paper/2301.11355/) (2023)
- [Data-Efficient Graph Grammar Learning for Molecular Generation](https://mlsanity.com/paper/2203.08031/) (2022)
- [LIDDIA: Language-based Intelligent Drug Discovery Agent](https://mlsanity.com/paper/2502.13959/) (2025)
- [Molecule Generation with Fragment Retrieval Augmentation](https://mlsanity.com/paper/2411.12078/) (2024)
- [Drug Discovery with Dynamic Goal-aware Fragments](https://mlsanity.com/paper/2310.00841/) (2023)
- [Generative Artificial Intelligence for Navigating Synthesizable Chemical Space](https://mlsanity.com/paper/2410.03494/) (2024)
- [POLO: Preference-Guided Multi-Turn Reinforcement Learning for Lead Optimization](https://mlsanity.com/paper/2509.21737/) (2025)

## Abstract (arXiv)

Molecule generation from extremely limited training data is a key challenge in drug discovery. Existing fragment-based methods are more suitable than atom-based approaches in this regime, but typically optimize fragment selection separately from downstream generation. Expert feedback is also especially valuable with limited data, yet translating such feedback into model objectives usually requires AI engineering expertise. We introduce FRAGMENTA, an end-to-end framework for small-data drug lead optimization with two components: (1) LVSEF, a fragment-based generator that jointly optimizes fragmentation and generation through a tabular reward-update mechanism, and (2) an agentic system that converts conversational expert feedback into updated generative objectives. Across three small-data datasets (11--104 molecules), LVSEF outperforms state-of-the-art methods in the smallest-data settings, matches them at larger scales, and trains ~16× faster. On three public protein targets, iterative closed-loop optimization improves final-round discovery yield by up to ~16% over one-shot LVSEF-only on kinase, with gains depending on how well feedback matches target chemistry. In a real-world cancer drug-discovery deployment, Human-Agent FRAGMENTA identified nearly twice as many molecules with favorable docking scores (< -6) as baseline methods.