# MoRA: On-the-fly Molecule-aware Low-Rank Adaptation Framework for LLM-based Multi-Modal Molecular Assistant

arXiv: https://arxiv.org/abs/2510.12245 · HTML: https://mlsanity.com/paper/2510.12245/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 43 references that are in the map.
- First posted on arXiv 14 Oct 2025.

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [HyperNetworks](https://mlsanity.com/paper/1609.09106/) (2016)
- [Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing](https://mlsanity.com/paper/2007.15779/) (2020)
- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [FLAVA: A Foundational Language And Vision Alignment Model](https://mlsanity.com/paper/2112.04482/) (2021)
- [SimVLM: Simple Visual Language Model Pretraining with Weak Supervision](https://mlsanity.com/paper/2108.10904/) (2021)
- [Multi-modal Molecule Structure-text Model for Text-based Retrieval and Editing](https://mlsanity.com/paper/2212.10789/) (2022)
- [A Molecular Multimodal Foundation Model Associating Molecule Graphs with Natural Language](https://mlsanity.com/paper/2209.05481/) (2022)
- [Unifying Molecular and Textual Representations via Multi-task Language Modelling](https://mlsanity.com/paper/2301.12586/) (2023)
- [InstructMol: Multi-Modal Integration for Building a Versatile and Reliable Molecular Assistant in Drug Discovery](https://mlsanity.com/paper/2311.16208/) (2023)
- [SELFIES and the future of molecular string representations](https://mlsanity.com/paper/2204.00056/) (2022)
- [Towards 3D Molecule-Text Interpretation in Language Models](https://mlsanity.com/paper/2401.13923/) (2024)
- [A Text-guided Protein Design Framework](https://mlsanity.com/paper/2302.04611/) (2023)
- [MolFM: A Multimodal Molecular Foundation Model](https://mlsanity.com/paper/2307.09484/) (2023)
- [MolTC: Towards Molecular Relational Modeling In Language Models](https://mlsanity.com/paper/2402.03781/) (2024)
- [LLaMo: Large Language Model-based Molecular Graph Assistant](https://mlsanity.com/paper/2411.00871/) (2024)
- [Large Language Models are In-Context Molecule Learners](https://mlsanity.com/paper/2403.04197/) (2024)
- [Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model](https://mlsanity.com/paper/2502.13449/) (2025)
- [PRESTO: Progressive Pretraining Enhances Synthetic Chemistry Outcomes](https://mlsanity.com/paper/2406.13193/) (2024)

## Built on by (most cited first, 1 of 1)

- [Graph is a Substrate Across Data Modalities](https://mlsanity.com/paper/2601.22384/) (2026)

## Abstract (arXiv)

Effectively integrating molecular graph structures with Large Language Models (LLMs) is a key challenge in drug discovery. Most existing multi-modal alignment methods typically process these structures by fine-tuning the LLM or adding a static adapter simultaneously. However, these approaches have two main limitations: (1) it optimizes a shared parameter space across all molecular inputs, limiting the model's ability to capture instance-specific structural features; and (2) fine-tuning the LLM for molecular tasks can lead to catastrophic forgetting, undermining its general reasoning capabilities. In this paper, instead of static task-oriented adaptation, we propose an instance-specific parameter space alignment approach for each molecule on-the-fly. To this end, we introduce Molecule-aware Low-Rank Adaptation (MoRA) that produces a unique set of low-rank adaptation weights for each input molecular graph. These weights are then dynamically injected into a frozen LLM, allowing the model to adapt its reasoning to the structure of each molecular input, while preserving the LLM's core knowledge. Extensive experiments demonstrate that on key molecular tasks, such as chemical reaction prediction and molecular captioning, MoRA's instance-specific dynamic adaptation outperforms statically adapted baselines, including a 14.1% relative improvement in reaction prediction exact match and a 22% reduction in error for quantum property prediction. The code is available at https://github.com/jk-sounds/MoRA.