MolEmb: Multimodal Large Language Models Can Be Strong Molecular Embedding Models

arXiv:2608.23646v1 Announce Type: new Abstract: 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 general molecular embedding models that produce embeddings conditioned on both a molecular profile and a natural-language semantic context. We introduce extbf{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 extbf{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.

Related

• A Large-Scale Dataset and Benchmark: Do Protein-Ligand Models Learn Binding Sites or Just Binding Likelihood?
• Rethinking Molecular Text Representations for LLMs: An Empirical Study
• C-MORAL: Controllable Multi-Objective Molecular Optimization with Reinforcement Alignment for LLMs
• SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing

Source: arXiv cs.AI | 2026-08-26