FORGE: Fragment-Oriented Ranking and Generation for Context-Aware Molecular Optimization

• URL: http://arxiv.org/abs/2605.10230v1
• Date: Mon, 11 May 2026 09:06:42 GMT
• Title: FORGE: Fragment-Oriented Ranking and Generation for Context-Aware Molecular Optimization
• Abstract summary: FORGE is a two-stage framework that reformulates molecular optimization as context-aware local editing.<n>Built on a compact 0.6B language model, FORGE consistently outperforms prior methods.<n>These results highlight the value of explicit fragment-level supervision as a more easily obtainable, scalable, and hallucination-less alternative to natural language training.
• Score: 38.801595580724616
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Molecular optimization seeks to improve a molecule through small structural edits while preserving similarity to the starting compound. Recent language-model approaches typically treat this task as prompt-conditioned sequence generation. However, relying on natural language introduces an inherent data-scaling bottleneck, often leads to chemical hallucinations, and ignores the strong context dependence of fragment effects. We present FORGE, a two-stage framework that reformulates molecular optimization as context-aware local editing. By utilizing automatically mined, verified low-to-high edit pairs instead of expensive human text annotations, Stage 1 ranks candidate fragments by their property contribution under the full molecular context to inject chemical prior, and Stage 2 generates explicit fragment replacements. Built on a compact 0.6B language model, FORGE further adapts to unseen black-box objectives through in-context demonstrations. Across Prompt-MolOpt, PMO-1k and ChemCoTBench, FORGE consistently outperforms prior methods, including substantially larger language models and graph methods. These results highlight the value of explicit fragment-level supervision as a more easily obtainable, scalable, and hallucination-less alternative to natural language training.

Related papers

• Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation [58.88304971824104]
  We introduce Fraglingo, an autoregressive fragment-based molecular generator.<n>Fraglingo predicts an attachment-aware fragment embedding and retrieves the next fragment.<n>It generalizes to fragment libraries up to 4x larger than those used during training without retraining.
  arXiv  Detail & Related papers  (2026-09-11T20:42:04Z)
• RTMol: Rethinking Molecule-text Alignment in a Round-trip View [4.597922051722059]
  We propose RTMol, a bidirectional alignment framework that unifies molecular captioning and text-to-SMILES generation through self-supervised round-trip learning.<n>Experiments demonstrate that RTMol enhances bidirectional alignment performance by up to 47% across various LLMs.
  arXiv  Detail & Related papers  (2025-11-15T09:55:55Z)
• Text-Guided Multi-Property Molecular Optimization with a Diffusion Language Model [20.250683535089617]
  We propose a text-guided multi-property molecular optimization method utilizing transformer-based diffusion language model (TransDLM)<n>By fusing physically and chemically detailed semantics with specialized molecular representations, TransDLM effectively integrates diverse information sources to guide precise optimization.
  arXiv  Detail & Related papers  (2024-10-17T14:30:27Z)
• FARM: Functional Group-Aware Representations for Small Molecules [55.281754551202326]
  We introduce Functional Group-Aware Representations for Small Molecules (FARM)<n>FARM is a novel model designed to bridge the gap between SMILES, natural language, and molecular graphs.<n>We evaluate FARM on the MoleculeNet dataset, where it achieves state-of-the-art performance on 11 out of 13 tasks.
  arXiv  Detail & Related papers  (2024-10-02T23:04:58Z)
• Expedited Training of Visual Conditioned Language Generation via Redundancy Reduction [61.16125290912494]
  $textEVL_textGen$ is a framework designed for the pre-training of visually conditioned language generation models. We show that our approach accelerates the training of vision-language models by a factor of 5 without a noticeable impact on overall performance.
  arXiv  Detail & Related papers  (2023-10-05T03:40:06Z)
• Translation between Molecules and Natural Language [43.518805086280466]
  We present a self-supervised learning framework for pretraining models on a vast amount of unlabeled natural language text and molecule strings. $textbfMolT5$ allows for new, useful, and challenging analogs of traditional vision-language tasks, such as molecule captioning and text-based de novo molecule generation.
  arXiv  Detail & Related papers  (2022-04-25T17:48:09Z)
• Molecular Attributes Transfer from Non-Parallel Data [57.010952598634944]
  We formulate molecular optimization as a style transfer problem and present a novel generative model that could automatically learn internal differences between two groups of non-parallel data. Experiments on two molecular optimization tasks, toxicity modification and synthesizability improvement, demonstrate that our model significantly outperforms several state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-11-30T06:10:22Z)
• Fragment-based Sequential Translation for Molecular Optimization [23.152338167332374]
  We propose a flexible editing paradigm that generates molecules using learned molecular fragments. We use a variational autoencoder to encode molecular fragments in a coherent latent space. We then utilize as a vocabulary for editing molecules to explore the complex chemical property space.
  arXiv  Detail & Related papers  (2021-10-26T21:20:54Z)
• Reinforced Molecular Optimization with Neighborhood-Controlled Grammars [63.84003497770347]
  We propose MNCE-RL, a graph convolutional policy network for molecular optimization. We extend the original neighborhood-controlled embedding grammars to make them applicable to molecular graph generation. We show that our approach achieves state-of-the-art performance in a diverse range of molecular optimization tasks.
  arXiv  Detail & Related papers  (2020-11-14T05:42:15Z)
• Graph Polish: A Novel Graph Generation Paradigm for Molecular Optimization [7.1696593196695035]
  We present a novel molecular optimization paradigm, Graph Polish, which changes molecular optimization from the traditional "two-language translating" task into a "single-language" task. We propose an effective and efficient learning framework T&S polish to capture the long-term dependencies in the optimization steps.
  arXiv  Detail & Related papers  (2020-08-14T08:36:13Z)
• A Deep Generative Model for Fragment-Based Molecule Generation [21.258861822241272]
  We develop a language model for small molecular substructures called fragments. In other words, we generate molecules fragment by fragment, instead of atom by atom. We show experimentally that our model largely outperforms other language model-based competitors.
  arXiv  Detail & Related papers  (2020-02-28T15:55:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.