Labs & Groups (1)
Models (5)
A protein language model that injects explicit structural constraints via structure-guided masking and a causal objective for efficient, structure-aware representation learning.
Scaling law study for protein language models that identifies compute-optimal training regimes for CLM and MLM architectures using 939M protein sequences.
A 100M-parameter foundation model trained on 50M+ human single-cell transcriptomic profiles, achieving state-of-the-art performance across diverse downstream tasks.
Unified 100-billion-parameter protein language model combining autoencoding and autoregressive objectives for protein understanding and generation.
Asymmetric encoder-decoder transformer for single-cell RNA-seq data that reduces FLOPs by 1-2 orders of magnitude while achieving state-of-the-art performance.