
Invalid SMILES Benefit Chemical Language Models: A Study
Skinnider (2024) shows that generating invalid SMILES actually improves chemical language model performance through …

Skinnider (2024) shows that generating invalid SMILES actually improves chemical language model performance through …

An end-to-end cheminformatics pipeline transforming 1D chemical formulas into 3D conformer datasets using graph …

Learn to implement VAEs in PyTorch: ELBO objective, reparameterization trick, loss scaling, and MNIST experiments on …

Supervised learning reveals hidden eigenvalue patterns that clustering missed, testing k-NN and logistic regression on …

Clustering analysis reveals why Coulomb matrix eigenvalues struggle with larger alkanes, using Dunn Index and silhouette …

Learn how dataset bias can lead to misleading results in NLP: a sarcasm detection model that actually learned to …

Explore molecular shape recognition using Coulomb matrix eigenvalues. Analysis of alkane isomers from data generation to …

Testing ML classification of congressional bills by policy area. Comparing Naive Bayes, Logistic Regression, and XGBoost …

Learn how Coulomb matrices encode 3D molecular structure for machine learning from basic theory to Python implementation …

A 47,000+ bill knowledge graph from Congress.gov with sponsor networks and 87% policy classification accuracy.

Perspective on SELFIES as a 100% robust SMILES alternative, with 16 future research directions for molecular AI.

Analytical derivation of Word2Vec's softmax objective factorization and a new framework for detecting semantic bias in …