Metraipolis › archive › dc_eadc859fbd1f
SafeMol: Dual-Modality Safety Alignment for Molecular Multimodal Models — arXiv:2609.33640
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33640
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:37:38.269Z
- SHA-256 of the archived text
57b6f6dbdf55b65ffc9f451ee60d0ffbd8d1b8da4082b923c57b24dea13d2315- Stable URL of this page
https://metraipolis.com/documento/dc_eadc859fbd1f
How to cite
SafeMol: Dual-Modality Safety Alignment for Molecular Multimodal Models — arXiv:2609.33640 — arXiv (en), https://arxiv.org/abs/2609.33640 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_eadc859fbd1f (Metraipolis archive dc_eadc859fbd1f, sha256 57b6f6dbdf55b65ffc9f451ee60d0ffbd8d1b8da4082b923c57b24dea13d2315)
BibTeX
@misc{dceadc859fbd1f,
title = {SafeMol: Dual-Modality Safety Alignment for Molecular Multimodal Models — arXiv:2609.33640},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33640},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_eadc859fbd1f; sha256 57b6f6dbdf55b65ffc9f451ee60d0ffbd8d1b8da4082b923c57b24dea13d2315},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33640. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.33640 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T22:14:28.003Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).