{"slug": "molemb-multimodal-large-language-models-can-be-strong-molecular-embedding-models", "title": "MolEmb: Multimodal Large Language Models Can Be Strong Molecular Embedding Models", "summary": "Researchers introduced MolEmb, a framework that adapts multimodal large language models (MLLMs) to generate molecular embeddings conditioned on both molecular structure and natural-language context, achieving competitive performance on molecular property prediction and cross-modal molecule–text retrieval. The team also introduced MolCAR, a diagnostic benchmark for context-aware retrieval, finding that context-aware embedding is primarily a data property of supervision. The findings suggest MLLMs can serve as general molecular embedding models for computational chemistry and drug discovery.", "body_md": "arXiv:2608.23646v1 Announce Type: new\nAbstract: Molecular embedding models can serve as foundational infrastructure for computational chemistry and drug discovery, where reusable vector representations support property prediction, virtual screening, and retrieval. Most molecular encoders are specialist models built around a single molecular view, producing unconditional vectors with no language interface for varying the representation. We ask whether multimodal large language models (MLLMs), which natively process images, text, and symbolic inputs, can instead serve as \\emph{general molecular embedding models} that produce embeddings conditioned on both a molecular profile and a natural-language semantic context. We introduce \\textbf{MolEmb}, a lightweight framework that adapts MLLMs by aligning molecular profiles with textual descriptions in a shared embedding space using a bidirectional contrastive objective. The resulting embedding model is competitive on molecular property prediction and supports cross-modal molecule--text retrieval in the same space. We further introduce \\textbf{MolCAR}, a diagnostic benchmark for context-aware retrieval, and find that context-aware molecular embedding is primarily a data property of the supervision. These results suggest that MLLMs are not merely chemistry assistants or generators, but a viable and extensible route to general molecular embedding models.", "url": "https://wpnews.pro/news/molemb-multimodal-large-language-models-can-be-strong-molecular-embedding-models", "canonical_source": "https://arxiv.org/abs/2608.23646", "published_at": "2026-08-26 04:00:00+00:00", "updated_at": "2026-08-26 04:15:31.544878+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research"], "entities": ["MolEmb", "MolCAR", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/molemb-multimodal-large-language-models-can-be-strong-molecular-embedding-models", "markdown": "https://wpnews.pro/news/molemb-multimodal-large-language-models-can-be-strong-molecular-embedding-models.md", "text": "https://wpnews.pro/news/molemb-multimodal-large-language-models-can-be-strong-molecular-embedding-models.txt", "jsonld": "https://wpnews.pro/news/molemb-multimodal-large-language-models-can-be-strong-molecular-embedding-models.jsonld"}}