Neural-Networks

Optimizing Sequence Models for Dynamical Systems

Ablation study deconstructing sequence models. Attention-augmented Recurrent Highway Networks outperform Transformers on …...

Summary of Kingma & Welling's foundational VAE paper introducing the reparameterization trick and variational …...

Summary of Burda, Grosse & Salakhutdinov's ICLR 2016 paper introducing Importance Weighted Autoencoders for tighter …...

The key difference between multi-sample VAEs and IWAEs: how log-of-averages creates a tighter bound on log-likelihood.

GTR-CoT uses graph traversal chain-of-thought reasoning to improve optical chemical structure recognition....

SubGrapher creates molecular fingerprints from images via functional group segmentation, enabling retrieval without full …...

αExtractor uses ResNet-Transformer to extract chemical structures from literature images, including noisy and hand-drawn …...

Clevert et al.'s two-stage CNN approach for converting molecular images to SMILES using CDDD embeddings and extensive …...

Chen et al.'s dual-stream encoder approach for robust molecular structure recognition from diverse real-world images …...

MolParser converts molecular images from scientific documents to machine-readable formats using E-SMILES....

Liu et al.'s ICLR 2025 paper introducing DenoiseVAE, which learns adaptive, atom-specific noise for better molecular …...

Lu et al. introduce SpaceFormer, a Transformer that models entire 3D molecular space (not just atoms) for superior …...