MoleculeNet introduces a large-scale benchmark suite for molecular machine learning, curating over 700,000 compounds across 17 datasets with standardized metrics, data splits, and featurization methods integrated into the DeepChem open-source library.
This study applies perplexity, a model-intrinsic metric from NLP, to rank de novo molecular designs generated by SMILES-based chemical language models and introduces a delta score to detect pretraining bias in transfer-learned CLMs.
This paper introduces ROGI-XD, a reformulation of the ROuGhness Index that enables fair comparison of QSPR surface roughness across molecular representations of different dimensionalities. Evaluating VAE, GIN, ChemBERTa, and ChemGPT representations, the authors show that pretrained chemical models do not produce smoother structure-property landscapes than simple molecular fingerprints or descriptors.
MOSES introduces a comprehensive benchmarking platform for molecular generative models, offering standardized datasets, evaluation metrics, and baselines. By providing a unified measuring stick, it aims to resolve reproducibility challenges in chemical distribution learning.
We explore the ‘Silent Failure’ mode of LLMs in production: the limits of 99% accuracy for reliability, how confidence decays in long documents, and why standard calibration techniques struggle to fix it.
This empirical study isolates the impact of chemical string representations on image-to-text translation models. It finds that while SMILES offers the highest overall accuracy, SELFIES provides a guarantee of structural validity, offering a trade-off for OCSR tasks.
Imago is an open-source, cross-platform C++ toolkit designed to recognize 2D chemical structure images from scientific papers and convert them into machine-readable molecule formats using a rule-based pipeline.
Performance evaluation of MolRec at the CLEF 2012 competition reveals a large performance gap between the automatic evaluation set (94-96% accuracy) and the manual evaluation set of complex patent structures (46-59% accuracy), with systematic analysis of failure modes including character grouping bugs, touching characters, and four-way junction vectorization.
Details the MolRec system for converting chemical diagram images into MOL files using vectorization, geometric rules, and graph construction. Achieved 95% accuracy on 1000 TREC 2011 benchmark images with comprehensive failure analysis of limitations.
A modern replication of Rahman’s 1964 argon molecular dynamics paper in LAMMPS, with a cached, vectorized Python analysis pipeline and an agreement table.
We explore LLM applications for page stream segmentation in insurance document processing, demonstrating that parameter-efficient fine-tuning achieves strong accuracy but revealing significant calibration challenges that limit deployment confidence.
What happens when you achieve 99.8% accuracy on sarcasm detection? You might have accidentally built a domain classifier. A cautionary ML tale about dataset bias.