InternLM Open-Sources Sage-SciPerception (ShuSheng·YuanZheng): Connecting Scientific Encoders to Language Models
InternLM has open-sourced Sage-SciPerception (ShuSheng·YuanZheng), which maps scientific encoders such as molecular graphs, spectra, crystal structures, protein and nucleic acid sequences into the language model input space through a trainable projector, and trains them separately by task. On QM9S joint spectral-structure recognition, Acc@1 is 49.75%; on BEACON ncRNA family classification, accuracy is 97.35%; and on JARVIS-DFT formation energy prediction, MAD/MAE is 15.03. The project provides three demos, and the projector checkpoint will be released on Hugging Face.
Body · AI translation
BranchesTagsOpen more actions menu
NameNameLast commit messageLast commit date
Repository files navigation
Sage-SciPerception
Sage Sci Perception · Shusheng·Yuanzheng
Scientific perception for language models.
English text is being prepared.
🤗 Models · GitHub · Demos · Results
Introduction
Sage Sci Perception (Shusheng·Yuanzheng) connects scientific encoders to a language model through trainable projectors. It maps molecular graphs, spectra, crystal structures, protein and nucleic-acid sequences, and scientific signals into the language model's input space, enabling natural-language analysis and structured scientific answers. Projectors are trained for each task.
Framework
Demos
Three spectra → molecular structure
IR, Raman and UV–Vis with streamed analysis and a recorded prediction.
Protein interface sites
Sequence positions and reference residue-level annotations.
Crystal structure and formation energy
Atomic geometry and a reference material property.
RNA interface refinement
Design revisions guided by experimental feedback.
Results
Selected results from task-specific checkpoints, evaluated on generated answers.
Benchmark Task Metric ↑ Score
QM9S Joint-spectrum structure identification Acc@1 (%) 49.75
JARVIS-DFT Formation-energy prediction MAD/MAE 15.03
MPBind, assembly-1 reconstruction PPI interface sites Mean per-protein F1 (%) 61.12
SciTS PHU01 ECG classification Macro-F1 (%) 71.53
SciTS NEU06 Sleep staging Macro-F1 (%) 72.01
BEACON ncRNA RNA family classification Accuracy (%) 97.35
Projector checkpoints will be available on Hugging Face.
AboutNo description, website, or topics provided.
ResourcesReadmeActivityCustom properties
Stars0 stars
Watchers0 watching
Forks0 forksReport repository