Search papers, labs, and topics across Lattice.
This study investigates the generalization capabilities of Large Language Models (LLMs) in the molecular domain by employing a Molecular Perturbation framework that generates syntax-valid structural variants of training molecules. The findings reveal that even minor structural edits can lead to significant performance declines on molecular tasks, indicating a limited local trust region and heightened sensitivity to changes. However, the use of In-Context Tuning (ICT) shows potential in expanding this trust region and enhancing robustness against structural variations, suggesting a viable approach for stabilizing molecular LLMs.
A single structural edit can drastically impair LLM performance in molecular tasks, highlighting the fragility of their generalization capabilities.
Large Language Models (LLMs) have recently shown promise in molecular discovery, yet a gap remains between their probabilistic nature over discrete sequential tokens and the rigid topological constraints of chemical space. This raises the question of whether molecular LLMs can generalize beyond the local neighborhoods induced by their sequence-based representations. To systematically investigate this question, we introduce a Molecular Perturbation framework that generates syntax-valid structural variants of training molecules under controlled Graph Edit Distance (GED) to probe the manifold regularity of molecular LLMs. Our analysis shows that even a single edit can cause substantial performance drops on common molecular tasks, revealing a narrow local trust region and fragile sensitivity to structural changes. Since similar molecules tend to exhibit similar properties, In-Context Tuning (ICT), which anchors predictions on structurally similar molecules, offers a natural way to mitigate such fragility. Our experiments also examine whether ICT confers robustness under controlled structural perturbations, and the results suggest that it can partially expand the local trust region and offer a promising direction for stabilizing molecular LLMs against structural variation.
NUS