Search papers, labs, and topics across Lattice.
This study fine-tunes the Uni-Mol2 molecular foundation model on the GS-LF benchmark for multi-label odor descriptor prediction and evaluates its performance across four diverse olfactory tasks without further deep-learning training. The results show that the fine-tuned model not only matches but often surpasses the performance of state-of-the-art olfaction-specific models, demonstrating effective transferability of learned representations. Notably, the model's ability to distinguish enantiomers highlights the advantages of three-dimensional molecular representations over traditional two-dimensional graph models, although challenges in predicting stereochemical perceptual outcomes remain.
Fine-tuning a single molecular foundation model can yield superior performance across multiple olfactory tasks, challenging the need for task-specific models.
Foundation models have transformed molecular property prediction, yet it remains unclear whether a molecular foundation model, fine-tuned on a single canonical olfactory prediction task, can learn representations that transfer across diverse machine olfaction problems. We investigate this question by fine-tuning Uni-Mol2 on the GS-LF benchmark for multi-label odor descriptor prediction and evaluating the resulting model, without additional deep-learning training, on four complementary downstream settings: cross-dataset odor descriptor prediction, odorous-versus-odorless classification, enantiomer evaluation, and odor mixture discriminability. The fine-tuned model matches or exceeds the performance of the state-of-the-art olfaction-specific baseline on the primary GS-LF benchmark and consistently transfers across these downstream evaluations. The enantiomer analysis further shows that three-dimensional molecular representations distinguish mirror-image molecules in a way that two-dimensional graph models fundamentally cannot, although accurately predicting the perceptual consequences of stereochemistry remains an open challenge. Together, these results support a train-once, transfer-across-tasks paradigm for machine olfaction and suggest that chemically pretrained molecular representations provide a strong foundation for transferable olfactory prediction.
