arXiv subjects

Sviatlana V. Lamaka

Publications and source records attributed to Sviatlana V. Lamaka.

3 recordsLinked to original sources

arXiv2026

Molecular Déjà Vu: Digit-Level Retrieval of Molecular Properties in Frontier Language Models

Large language models (LLMs) are increasingly employed to predict molecular properties. However, prediction error alone cannot distinguish prediction from retrieval of published values. We audit 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. Significant retrieval is concentrated on 5 datasets, with isolated flagged LLMs elsewhere. Increasing the reasoning setting raises the number of flagged model--dataset combinations from 47 to 89 of 264. An in-context blinding experiment reduces retrieval but leaves a quarter of the combinations flagged. Blinding changes model rankings and increases errors. Because blinding also removes chemically interpretable structure, the error increase can only be partially attributed to reduced retrieval.

cs.AI↗

arXiv2026

In-Context Molecular Property Prediction with LLMs: A Blinding Study on Memorization and Knowledge Conflicts

The capabilities of large language models (LLMs) have expanded beyond natural language processing to scientific prediction tasks, including molecular property prediction. However, their effectiveness in in-context learning remains ambiguous, particularly given the potential for training data contamination in widely used benchmarks. This paper investigates whether LLMs perform genuine in-context regression on molecular properties or instead rely on verbatim retrieval of memorized target values. Furthermore, we analyze the interplay between pre-trained knowledge and in-context information through a series of progressively blinded experiments. We evaluate nine LLM variants across three families (GPT-4.1, GPT-5, Gemini 2.5) on three MoleculeNet datasets (Delaney solubility, Lipophilicity, QM7 atomization energy) using a systematic blinding approach that iteratively reduces available information, complemented by 0-, 60-, and 1000-shot in-context sample sizes as an additional control for information access. To validate the memorization analysis and the blinding experiments, we add a positive and a negative control for the memorization experiments and structural reference baselines for the multi-shot experiments as well as bootstrap confidence intervals for all results. We find no evidence of verbatim retrieval on the legacy benchmarks and show that blinding exposes conflicts between pre-trained knowledge and in-context information. This work provides a principled framework for evaluating molecular property prediction under controlled information access.

cs.LG↗

arXiv2021

Computational modeling of degradation process of biodegradable magnesium biomaterials

Despite the advantages of using biodegradable metals in implant design, their uncontrolled degradation and release remain a challenge in practical applications. A validated computational model of the degradation process can facilitate the tuning of implant biodegradation by changing design properties. In this study, a physicochemical model was developed by deriving a mathematical description of the chemistry of magnesium biodegradation and implementing it in a 3D computational model. The model parameters were calibrated using the experimental data of hydrogen evolution by performing a Bayesian optimization routine. The model was validated by comparing the predicted change of pH in saline and buffered solutions with the experimentally obtained values from corrosion tests, showing maximum 5% of difference, demonstrating the model's validity to be used for practical cases.

cs.CE↗