This 1992 paper introduces Kekulé, one of the first complete Optical Chemical Structure Recognition (OCSR) systems. It details a pipeline integrating raster-to-vector conversion, neural network-based OCR, and rule-based logic to convert printed chemical diagrams into connection tables.
A 1994 paper identifying why standard least-squares networks fail at inverse problems (multi-valued mappings). It introduces the Mixture Density Network (MDN), which predicts the parameters of a Gaussian Mixture Model to capture the full conditional probability density.
This paper describes an early prototype system that digitizes chemical structure diagrams from scanned documents. It employs a multi-stage pipeline involving convex bounding polygon extraction, vectorization, and rule-based heuristics to generate MDL Molfiles.
This paper presents a 4-variable kinetic model coupling surface reaction dynamics with structural phase transitions to reproduce complex oscillatory behavior on Pt(110).
This methodological paper proposes a comprehensive pipeline to digitize chemical structure images. It achieves 97% reconstruction accuracy on benchmarks by combining a topology-preserving vectorizer with a chemical knowledge validation module.
Proposes the Hyperbolic Algorithm for Euclidean field theory simulations. By adding a second-order fictitious time derivative to the Langevin equation, the method reduces systematic errors from O(ε) down to O(ε²).
Stillinger and Weber propose a 3-body interaction potential that stabilizes the diamond crystal structure of silicon and reproduces liquid properties through molecular dynamics, addressing the inability of standard pair potentials to model tetrahedral semiconductors.
This work validated classical Molecular Dynamics for simulating liquids, revealing the ‘cage effect’ in velocity autocorrelation and establishing predictor-corrector integration algorithms for N-body problems.
Kingma and Welling’s 2013 paper introducing Variational Autoencoders and the reparameterization trick, enabling end-to-end gradient-based training of generative models with continuous latent variables by moving the stochasticity outside the computational graph so that gradients can flow through a deterministic path.
Burda et al.’s ICLR 2016 paper introducing Importance Weighted Autoencoders, which use importance sampling to derive a strictly tighter log-likelihood lower bound than standard VAEs, addressing posterior collapse and improving generative quality. The model architecture remains the same.
The 2020 paper that introduced SELFIES: Mario Krenn and colleagues created a molecular representation that solves SMILES validity problems. It guarantees every generated string corresponds to a valid chemical structure.
David Weininger introduced SMILES notation in 1988, establishing encoding rules for representing chemical structures as compact, human-readable strings.