All Competitors

Conditional residual-diffusion model for multi-parametric brain MRI that synthesizes a missing contrast and returns pixel-level uncertainty maps.

Protein dynamics prediction from one equilibrium structure, giving per-residue directional flexibility and pairwise residue couplings.

3' UTR foundation model pretrained on 7 million eukaryotic untranslated regions for mRNA stability prediction and de novo UTR sequence design.

Spatial RNA-to-protein prediction turning transcriptomics-only sections into virtual protein maps via two adaptively reweighted graph channels.

RNA 3D structure prediction from a single sequence, pairing a frozen language model with base-pair probabilities in a compact folding trunk.

Self-supervised encoder for tandem mass spectra that learns proteomics representations from 1.63 billion MS/MS scans with no peptide labels.

Crop phenotyping vision-language model that answers questions about arbitrarily shaped image regions, routing each task to its own expert branch.

Pocket-conditioned 3D ligand generation for structure-based drug design, with variance-exploding diffusion that keeps generated geometry low-strain.

Structure-based protein-RNA screening that reads position-wise nucleotide preferences off a single complex structure and ranks libraries in seconds.

Retention time prediction for reversed-phase HPLC, trained on 179 chromatographic methods plus MD and DFT properties computed for 21,431 molecules.

MSA-free all-atom protein structure prediction, folding monomers from the representations of a 3B-parameter metagenomic protein language model.

Nuclear segmentation and subtype classification in immunohistochemistry slides, steering frozen SAM and UNI encoders with frequency-domain prompts.

Unified neural foundation model jointly pretrained on intracranial EEG and intracortical spiking activity for motor and language decoding.

Plant genomic foundation models on a bidirectional Mamba-2 backbone, 318M to 3.2B parameters, pretrained on 320 species with context out to 128 kbp.

Chest radiograph foundation model with a single latent state behind both diagnostic readout and report-conditioned image generation.

Capsid-specialized protein language model that finds divergent viral capsid proteins in metagenomes and encodes shell architecture from sequence.

Protein inverse folding by Dirichlet flow matching, refining the whole sequence at once with fixed-residue inpainting and soft residue priors.

Sequence-only protein-RNA interaction model that predicts binding affinity, maps residue- and nucleotide-level interfaces, and designs binding RNAs.

All-atom diffusion model for joint protein, peptide, nucleic acid and ligand complex prediction, scored on interface accuracy and steric validity.

Cortical surface registration network that learns which folds predict functional organization, using multi-modal supervision during training only.

Sleep foundation model whose frozen encoder scores polysomnography events, predicts incident disease, and transfers to wearable EEG and actigraphy.

Protein-sequence-conditioned mixture-of-experts language model that designs small molecules and emits an explicit synthesis route for each one.

Diffusion generator for 2D molecular graphs that blends spatial and spectral denoising branches under signal-to-noise-guided fusion weights.