Reproduce “RIBOSPAN: A Long-Context RNA Foundation Model for Versatile RNA Modeling” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ziyuan Wang, Bohao Tang, Fei Zhang, Shuo Han, Pengfei Liu
RIBOSPAN is a 1.61-billion-parameter bidirectional Transformer pretrained on long RNA transcripts with one token per nucleotide and context up to 10,240 nucleotides. It uses native long-context training rather than inference-time extension. In a rearrangement benchmark, it has lower cross-region similarity and distal diffusion than a YaRN-extended short-context model, but lower context separation; these measures assess representations, not biological function.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.