Molecular D\'ej\`a Vu: Digit-Level Retrieval of Molecular Properties in Frontier Language Models
This work audits 22 frontier models on 12 molecular regression datasets in a zero-shot setting, assessed against a molecule-blind reference derived from each dataset's labels, finding significant retrieval is concentrated on 5 datasets, with isolated flagged LLMs elsewhere.
Matthias Busch, M. Tacke, Sviatlana V. Lamaka et al.
· 0 citations