← Back to all news
arXiv AI By Matthias Busch, Marius Tacke, Sviatlana V. Lamaka, Mikhail L. Zheludkevich, Christian J. Cyron, Roland C. Aydin, Christian Feiler

Molecular D\'ej\`a Vu: Digit-Level Retrieval of Molecular Properties in Frontier Language Models

Read the original on arXiv AI →

The paper evaluates 22 state‑of‑the‑art large language models on 12 molecular regression datasets in a zero‑shot setting, comparing their predictions to a molecule‑blind reference derived from the datasets’ labels. It finds that many models retrieve published values rather than truly predicting properties, with significant retrieval concentrated on five datasets and additional flagged instances elsewhere. Increasing the reasoning setting doubles the number of flagged model–dataset pairs, and an in‑context blinding experiment reduces but does not eliminate retrieval, also altering model rankings and increasing errors.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Aug 5

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

arXiv:2603. 25857v3 Announce Type: replace Abstract: The capabilities of large language models (LLMs) have expanded beyond natural language processing to scientific prediction tasks, including molecular property prediction.

By Matthias Busch, Marius Tacke, Sviatlana V. Lamaka, Mikhail L. Zheludkevich, Christian J. Cyron, Christian Feiler, Roland C. Aydin
llmsnlpbenchmarks
More like this →
arXiv AI
Sep 7

Molecular D\'ej\`a Vu: Digit-Level Retrieval of Published Values in Frontier Language Models

The paper audits 22 frontier language models on 12 molecular property regression benchmarks to assess verbatim retrieval of published values. It finds widespread but benchmark‑specific retrieval, with over 50% of models retrieving exact values on five datasets and isolated occurrences on others. Experiments at different reasoning levels show that higher reasoning increases retrieval flags, and attempts to interrupt retrieval reveal that top models can still recognize transformed SMILES and original labels. Suppressing retrieval reduces prediction error variance, indicating that predictive performance is not solely due to memorized values.

By Matthias Busch, Marius Tacke, Sviatlana V. Lamaka, Mikhail L. Zheludkevich, Christian J. Cyron, Roland C. Aydin, Christian Feiler
llmsbenchmarks
More like this →
arXiv Machine Learning
Sep 22

MolSC: Leveraging Substituent Contributions to Enhance Fine-grained Molecular Understanding in LLMs

MolSC is a new dataset of 181,000 substituent-level examples that captures how attaching specific substituents to molecular scaffolds changes properties such as bioactivity and physicochemical descriptors. The authors also provide MolSC-Bench, a held‑out benchmark of 1,541 examples that are disjoint from MolSC at scaffold, substituent, and molecule levels. Experiments show that training molecular large language models on MolSC markedly improves their ability to predict substituent contributions, outperforming existing models on a range of downstream chemistry tasks.

By Hyuntae Park, Sooyeon Kim, Jiwon Park, SangKeun Lee
llmsnlpbenchmarks
More like this →
arXiv Machine Learning
Jun 5

MolE-RAG: Molecular Structure-Enhanced Retrieval-Augmented Generation for Chemistry

arXiv:2606. 05693v1 Announce Type: new Abstract: Large language models (LLMs) have shown promise for molecular property prediction, but their ability to reason over chemical structures remains limited, as molecular representations such as SMILES differ substantially from the natural language on which LLMs are primarily trained.

By Joey Chan, Wonbin Kweon, Ashley Shin, Niharika Bhattacharjee, Pengcheng Jiang, Yue Guo, Jiawei Han
llmsragfine-tuning
More like this →