Stars
verl: Volcano Engine Reinforcement Learning for LLMs
RAGEN leverages reinforcement learning to train LLM reasoning agents in interactive, stochastic environments.
Derivative-Free Guidance in Diffusion Models with Soft Value-Based Decoding. For controlled generation in DNA, RNA, proteins, molecules (+ images)
Analyzing Hacker News discussions from a decade ago in hindsight with LLMs
Github repository for nanoSPLITS manuscript data and R scripts
Analysis and method implementation for exploring private information leakage from single-cell RNA-seq count matrices
Latent Collaboration in Multi-Agent Systems
deep residual neural network for classifying the pathogenicity of missense mutations.
A foundation model for spatial transcriptomics. It generates contextual gene representations within single cell and spatial niche.
FlashRNA - An Efficient Model for Regulatory Genomics
Context-Aware Regularization with Markovian Integration for Attention-Based Nucleotide Analysis [NeurIPS2025]
DNA sequences in a multiple sequence alignment transformer.
A family of codon-resolution language models trained on 130 million protein-coding sequences from over 20,000 species.
OligoGym is a python package that streamlines processes involving featurization, training and evaluation of predictive models of oligonucleotide properties. Oligonucleotides include antisense oligo…
DSPy: The framework for programming—not prompting—language models
An API standard for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym)
Experimental design and (multi-objective) bayesian optimization.
With this snakemake pipeline you can process your MPRA sequencing data (assignment and count). It is the standard MPRA pipeline of the IGVF consortium and a further development of MPRAflow.
A portable, flexible, parallelized tool for complete processing of massively parallel reporter assay data
Multi-task and masked language model-based protein sequence embedding models.
Training setup for Langchain's Open Deep Research
Methods for efficiently and accurately fitting deep models on public phenotype data.