# SafeMol: Dual-Modality Safety Alignment for Molecular Multimodal Models

arXiv: https://arxiv.org/abs/2609.33640 · HTML: https://mlsanity.com/paper/2609.33640/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 32 of its 38 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Safety alignment and jailbreak attacks](https://mlsanity.com/alignment/safety-alignment-and-jailbreak-attacks/)
- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 30 of 32)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Learning Transferable Features with Deep Adaptation Networks](https://mlsanity.com/paper/1502.02791/) (2015)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts](https://mlsanity.com/paper/2311.05608/) (2023)
- [MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models](https://mlsanity.com/paper/2311.17600/) (2023)
- [OR-Bench: An Over-Refusal Benchmark for Large Language Models](https://mlsanity.com/paper/2405.20947/) (2024)
- [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)
- [MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter](https://mlsanity.com/paper/2310.12798/) (2023)
- [How Does Information Bottleneck Help Deep Learning?](https://mlsanity.com/paper/2305.18887/) (2023)
- [InstructMol: Multi-Modal Integration for Building a Versatile and Reliable Molecular Assistant in Drug Discovery](https://mlsanity.com/paper/2311.16208/) (2023)
- [SPA-VL: A Comprehensive Safety Preference Alignment Dataset for Vision Language Model](https://mlsanity.com/paper/2406.12030/) (2024)
- [Towards 3D Molecule-Text Interpretation in Language Models](https://mlsanity.com/paper/2401.13923/) (2024)
- [ETA: Evaluating Then Aligning Safety of Vision Language Models at Inference Time](https://mlsanity.com/paper/2410.06625/) (2024)
- [Prot2Text: Multimodal Protein's Function Generation with GNNs and Transformers](https://mlsanity.com/paper/2307.14367/) (2023)
- [Unraveling and Mitigating Safety Alignment Degradation of Vision-Language Models](https://mlsanity.com/paper/2410.09047/) (2024)
- [LLaMo: Large Language Model-based Molecular Graph Assistant](https://mlsanity.com/paper/2411.00871/) (2024)
- [ReactXT: Understanding Molecular "Reaction-ship" via Reaction-Contextualized Molecule-Text Pretraining](https://mlsanity.com/paper/2405.14225/) (2024)
- [Cross-Modal Safety Mechanism Transfer in Large Vision-Language Models](https://mlsanity.com/paper/2410.12662/) (2024)
- [VLM-Guard: Safeguarding Vision-Language Models via Fulfilling Safety Alignment Gap](https://mlsanity.com/paper/2502.10486/) (2025)
- [Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model](https://mlsanity.com/paper/2502.13449/) (2025)
- [ChemSafetyBench: Benchmarking LLM Safety on Chemistry Domain](https://mlsanity.com/paper/2411.16736/) (2024)
- [Learning Optimal Multimodal Information Bottleneck Representations](https://mlsanity.com/paper/2505.19996/) (2025)
- [Dynamic Multimodal Information Bottleneck for Multimodality Classification](https://mlsanity.com/paper/2311.01066/) (2023)
- [Adversary-Aware DPO: Enhancing Safety Alignment in Vision Language Models via Adversarial Training](https://mlsanity.com/paper/2502.11455/) (2025)
- [How Does Vision-Language Adaptation Impact the Safety of Vision Language Models?](https://mlsanity.com/paper/2410.07571/) (2024)
- [SMILES-Prompting: A Novel Approach to LLM Jailbreak Attacks in Chemical Synthesis](https://mlsanity.com/paper/2410.15641/) (2024)
- [Omni-Mol: Multitask Molecular Model for Any-to-any Modalities](https://mlsanity.com/paper/2502.01074/) (2025)

## Abstract (arXiv)

Molecular multimodal models support diverse understanding and generation tasks but may introduce safety vulnerabilities when handling hazardous molecules. In this work, We reveal substantial jailbreak vulnerabilities under both text-only and graph-conditioned settings. Our analysis further shows that safety robustness must hold across input modalities while balancing safety, over-refusal, and utility. To address these challenges, we construct SafeMolBench, a molecular multimodal safety-alignment benchmark with 3702 samples covering 618 unique hazardous molecules and safe molecular tasks, organized into hazardous-harmful, hazardous-allowed, and utility-replay subsets to support unified training and evaluation of safety, over-refusal, and utility. Based on SafeMolBench, we propose SafeMol, a parameter-efficient safety alignment framework that jointly optimizes lightweight modules across text-only and graph-conditioned inputs, uses MMD for distribution-level representation alignment to reduce modality-induced discrepancies, and explicitly models molecular hazardousness and harmful operational intent. Experiments on SafeMolBench show that SafeMol reduces attack success by several tens of percentage points while largely maintaining low over-refusal and preserving molecular-task utility.