Papers › Mol-LLM: Multimodal Generalist Molecular LLM with Improved Graph Utilization

Mol-LLM: Multimodal Generalist Molecular LLM with Improved Graph Utilization

5 Feb 2025arXiv:2502.02810archive 2025-07-28

Chanhui Lee, Hanbum Ko, Yuheon Song, Yongjun Jeong, Rodrigo Hormazabal, Sehui Han, Kyunghoon Bae, Sungbin Lim, Sungwoong Kim

Recent advances in large language models (LLMs) have led to models that tackle diverse molecular tasks, such as chemical reaction prediction and molecular property prediction. Large-scale molecular instruction-tuning datasets have enabled sequence-only (e.g., SMILES or SELFIES) generalist molecular LLMs, and researchers are now exploring multimodal approaches that incorporate molecular structural information for further gains. However, a genuinely multimodal, generalist LLM that covers a broad spectrum of molecular tasks has yet to be fully investigated. We observe that naive next token prediction training ignores graph-structural information, limiting an LLM's ability to exploit molecular graphs. To address this, we propose (i) Molecular structure Preference Optimization (MolPO), which facilitates graph usage by optimizing preferences between pairs of correct and perturbed molecular structures, and (ii) an advanced graph encoder with a tailored pre-training strategy to improve the effect of graph utilization by MolPO. Building on these contributions, we introduce Mol-LLM, the first multimodal generalist model that (a) handles a broad spectrum of molecular tasks among molecular LLMs, (b) explicitly leverages molecular-structure information, and (c) takes advantage of extensive instruction tuning. Mol-LLM attains state-of-the-art or comparable results across the most comprehensive molecular-LLM benchmark-even on out-of-distribution datasets for reaction and property prediction, where it surpasses prior generalist molecular LLMs by a large margin.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Chemical Reaction PredictionMolecular Property PredictionMolecule CaptioningPredictionProperty Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Molecule Captioning ChEBI-20 Mol-LLM (SELFIES) BLEU-2 58.7 #17 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM (SELFIES) BLEU-4 51.5 #17 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM (SELFIES) METEOR 61.7 #17 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM (SELFIES) ROUGE-1 62.7 #17 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM (SELFIES) ROUGE-2 48.7 #17 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM (SELFIES) ROUGE-L 57.1 #17 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM BLEU-2 56.0 #22 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM BLEU-4 49.0 #22 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM METEOR 59.3 #22 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM ROUGE-1 52.4 #22 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM ROUGE-2 37.0 #22 of 33 Archive leaderboard report
Molecule Captioning ChEBI-20 Mol-LLM ROUGE-L 46.7 #22 of 33 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections