Shanghai AI Lab releases Sage-SciPerception to map scientific data into language models

Official source · auto-listedOriginal title internlm/Sage-SciPerception

Shanghai AI Lab (InternLM) published Sage-SciPerception on Hugging Face. The project connects scientific encoders to a language model through trainable projectors, mapping molecular graphs, spectra, crystal structures, protein and nucleic-acid sequences, and scientific signals into the language model's input space for natural-language analysis and structured scientific answers. Projectors are organized by language backbone, modality, and task. Selected task-specific checkpoint results on the model card include 49.75% Acc@1 on QM9S joint-spectrum structure identification and 97.35% accuracy on BEACON ncRNA family classification.

Auto-listed by dalao.ai: the headline is a fixed template and the original title is shown as written. The official page is the record; its publisher keeps authorship.